On February 19th, the birthday of Nicolaus Copernicus, Professor James G. Fujimoto was awarded the Honorary Doctorate Degree at the Nicolaus Copernicus University in Poland. Professor Fujimoto was awarded the University’s highest academic distinction, Doctor Honoris Causa, by Rector Andrzej Tretyn and Dean Andrzej Kowalczyk in recognition of his contribution to the fields of biomedical optics and medicine, his training and mentoring of young scientists, and his service to the international scientific community.

Read more in the February 20, 2015 Research Laboratory of Electronics article, also posted below.

On February 19th, the birthday of Nicolaus Copernicus, Professor James G. Fujimoto was awarded the Honorary Doctorate Degree at the Nicolaus Copernicus University in Poland. Professor Fujimoto was awarded the University’s highest academic distinction, Doctor Honoris Causa, by Rector Andrzej Tretyn and Dean Andrzej Kowalczyk in recognition of his contribution to the fields of biomedical optics and medicine, his training and mentoring of young scientists, and his service to the international scientific community.

The Copernicus University, located in Torun, Poland has over 30,000 students and includes faculties of Economics, Education, Law, Medicine, Political Science, and Physics as well as numerous faculties in arts and sciences. Nicolaus Copernicus’ work De revolutionibus orbium coelestium (On the Revolutions of the Celestial Spheres) described the helio-centeric solar system and is considered a key result in the history of science. His vision led to a profound change in the perception of mankind’s place in the universe, while reinforcing the critical role of scientific and analytical methodology. Without careful scientific methodology or systematic analysis, first perceptions and beliefs are often incorrect. Copernicus was a multidisciplinary person, astronomer, mathematician, physician, literary scholar, economist, diplomat and expert on Canon law. His broad, multidisciplinary training enabled him to transcend the historical limits of vision and correct a fundamental error in mankind’s perception of reality.

In his commencement speech, “The limits of vision – and how we might transcend them”, Prof. Fujimoto commented that the high degree of specialization in modern science, technology, medicine and society requires years of focused training which can limit of vision and lead us to a narrow perception of reality. Progress has been so great that it is now almost impossible for a single person to have the depth and breadth of knowledge that Copernicus did. In order to extend the boundaries of knowledge, advance health care, create new economic opportunities, or make other positive contributions to society, it is increasingly necessary to work in multidisciplinary teams. Not only is it critical to work collaboratively, but it is also important to develop an understanding of the how other disciplines perceive reality, approach problems and create solutions. Current biomedical research represents an example of how multidisciplinary approaches can lead to powerful advances in patient care through collaboration and interaction among scientists, engineers, clinicians, business and government.

Professor James G. Fujimoto obtained his bachelors, masters, and doctorate from the Massachusetts Institute of Technology in 1979, 1981, and 1984 respectively. He performed his doctoral studies under the supervision of Professor Erich Ippen in ultrafast optics. Since 1985 Dr. Fujimoto has been in the Department of Electrical Engineering and Computer Science and Research Laboratory of Electronics at MIT where he is currently Elihu Thomson Professor of Electrical Engineering. Dr. Fujimoto’s group (Biomedical Optical Imaging and Biophotonics Group) and collaborators were responsible for the invention and development of optical coherence tomography (OCT). His group’s landmark paper, “Optical Coherence Tomography,” which appeared in Science in 1991, ushered in a new era in clinical biophotonic imaging and has remained one of the highest cited papers in the field of biophotonics. Dr. Fujimoto and his collaborators have played crucial roles in many of the major innovations in the OCT field. OCT is now standard clinical imaging modality in ophthalmology for the detection and treatment monitoring of macular degeneration, diabetic retinopathy and glaucoma. There are an estimated 20–30 million ophthalmic imaging procedures performed worldwide every year. There are hundreds of researchers internationally working on OCT in such diverse fields as cardiology, endoscopy and cancer surgery. Last year, the global sales of OCT systems exceeded $400 million and there are more than 36 OCT systems companies.

Dr. Fujimoto has been influential as an educator and has trained numerous researchers who became leaders in the fields of photonics and biomedicine. He is also active in scientific service, having served as co-chair of international meetings such as the Conference on Lasers and Electro Optics, the European Conferences on Biomedical Optics, and Ultrafast Phenomena. Since 2003 Dr. Fujimoto has served as co-chair of the SPIE Biomedical Optics symposium, the largest international meeting on biophotonics. He also served as a Director of the Optical Society of America from 2000 to 2003 and is currently a Director of the SPIE the International Society for Optics and Photonics.

Dr. Fujimoto received the Discover Magazine Award for Technological Innovation in medical diagnostics in 1999, was co-recipient of the Rank Prize in Optoelectronics in 2002, received the Zeiss Research Award in 2011 and was co-recipient of the Champalimaud Vision Prize in 2012. He was elected to the National Academy of Engineering, American Academy of Arts and Sciences, and National Academy of Sciences. Dr. Fujimoto is also a Fellow of the OSA, APS, and IEEE.

Working with Mr. Eric Swanson, Dr. Fujimoto was a co-founder of the startup company Advanced Ophthalmic Devices, which developed OCT for ophthalmic imaging and was acquired by Carl Zeiss. He and Mr. Swanson also co-founded LightLab Imaging, which developed cardiovascular OCT and was acquired by Goodman, Ltd and St. Jude Medical.

Erna Viterbi, a warm and gracious philanthropist who with her husband, Qualcomm co-founder Andrew ’56, SM ’57, gave generously to MIT and a variety of other institutions, died Feb. 17 in San Diego.

“In the long adventure of their lives together, Erna and Andrew were terrific partners — curious, generous, thoughtful, and creative in everything they did,” MIT President L. Rafael Reif says. “Erna brightened every encounter and enriched every conversation. It was a delight to know her, and I am deeply grateful for her friendship and for everything she did for MIT.”

At MIT, the Viterbis established endowed professorships and fellowships for graduate students in the departments of electrical engineering and computer science and biological engineering. Four faculty members and more than 40 students on campus have benefited directly from their generosity. Together with her husband, Viterbi has also given generously to undergraduate scholarships to help MIT maintain its need-blind admissions policy. [Photo above: Andrew and Erna Viterbi attended a dinner at MIT last year to celebrate faculty who have held or are holding Viterbi chairs. Seated (l-r): Viterbi Professor Dina Katabi, Andrew Viterbi, Erna Viterbi, Viterbi Professor Scott Manalis. Standing (l-r): Douglas Lauffenburger, head of the Department of Biological Engineering; Ronald Rivest, who held the first Viterbi Professorship; Viterbi Professor Daniela Rus; Dean of Engineering Ian Waitz; Anantha Chandrakasan, head of the Department of Electrical Engineering and Computer Science; and Eric Grimson, Chancellor for Academic Advancement.  Photo: Rose Lincoln, courtesy MIT News/Resource Development]

“We are deeply sorrowful for Erna’s passing,” says Douglas A. Lauffenburger, the Ford Professor of Bioengineering and head of the Department of Biological Engineering. “She had a magnificent spirit and was a wonderful partner with Andrew in their extraordinary support for the faculty and students of our MIT biological engineering program.”

“Andrew and Erna Viterbi have given us a tremendous opportunity to honor the very best faculty and graduate students,” adds Anantha P. Chandrakasan, the Joseph F. and Nancy P. Keithley Professor of Electrical Engineering and head of the Department of Electrical Engineering and Computer Science. “Their support also allows our faculty and students to explore new research directions that are perhaps not as easy to fund in other ways.”

Daniela Rus, one of three current Andrew and Erna Viterbi Professors at MIT, says the family’s support has helped her to explore new ideas at the intersection of communication, control, and computation. “Erna was always so warm. I am so grateful I had a chance to know her,” says Rus, who is also the director of MIT’s Computer Science and Artificial Intelligence Laboratory.

The Viterbis have also been involved in other aspects of MIT community life. For example, last November the two attended MIT Hillel’s “Leading Jewish Minds @ MIT” faculty luncheon program, at which Andrew Viterbi spoke about the evolution of technology over his career; the couple had the opportunity to interact with the many Viterbi scholars and fellows who attended the event. “Erna was such a gracious person,” remembers Rabbi Michelle H. Fisher, executive director of MIT Hillel, which the Viterbis have also helped to support. “At another event I attended with her, I remember how interested she was both in sharing her own story and in hearing mine.”

According to published accounts of her life, Erna Finci Viterbi was born in Sarajevo, a descendant of Sephardic Jews who were expelled from Spain. In 1941, during World War II, the Finci family fled German-occupied Yugoslavia for the Italian-occupied zone from which they were deported and interned in the Parma region of Italy. In 1943, when the Nazis occupied Italy, the family was saved from deportation to extermination camps by the people of Gramignazzo di Sissa, the village where they had been interned. Other Italians helped them escape to Switzerland, where they waited out the war.

In 1950, the Finci family resettled in California, where Erna met Andrew Viterbi; the two were married in 1958. “She became his equal lifetime partner, sharing in all major decisions and she was usually by his side as he scribbled notes on communication theory at home or at family gatherings,” according to the Andrew J. and Erna Viterbi Family Archives at the University of Southern California.

Erna Viterbi is survived by her husband, Andrew; son, Alan; daughter, Audrey; and five grandchildren.

THe MIT News Office has featured EECS senior Sheldon Trotman for his innovations directed to solving inefficiencies.

Read more in the February 26, 2015 MIT News Office article by Julia Sklar, MIT News correspondent titled "Fighting inefficiencies - Senior Sheldon Trotman designs computer programs to streamline human behavior," also posted below.

When MIT senior Sheldon Trotman walks into any room, he almost instinctively looks for inefficiencies. The electrical engineering and computer science major is bent on streamlining our world, and has already founded several small companies that aim to do so.

Even while meeting at a small coffee shop, he is quick to point out a simple fix to a problem that the shop’s proprietors may not even know they have: Many restaurants and workplaces with shift-based scheduling rely on clunky spreadsheets, email lists, or phone calls to coordinate employees who pick up and drop shifts at the last minute. Trotman has designed an app that would let workers do all of this remotely. “You could go on your phone and say, ‘Hey, I don’t want to work today,’ and it gets sent out to all your co-workers and your manager, instead of having to pick up the phone and call everyone individually, or clogging up their email inbox,” Trotman says. “I just see inefficiencies everywhere, and I want to fix them.”

Looking for a connection

Trotman grew up primarily in Maryland, and attended a magnet middle school that focused on math and science, providing him with resources in computer science. But his initial interest in technology really stemmed from his efforts to fit in.

“I did track, football, and chess, which was a pretty strange combination in a cliquey public school,” Trotman says. “I was also the only person of color [in that middle school program], and that can be isolating.”

Looking for a way to connect, at an early age he became hooked on teaching himself the basics of artificial intelligence, so that he could someday create something that could communicate back and forth.

At MIT, Trotman’s first research experience was in the Humans and Automation Laboratory, where he assisted in simulating human interactions with autonomous vehicles. His work supported the research of then-associate professor of aeronautics and astronautics Missy Cummings and graduate student Jason Ryan.

“I’m definitely still stuck on artificial intelligence,” Trotman says. “It’s just another way to make people more effective at what they’re already good at doing: unstructured problem solving.”

Getting down to business

But Trotman realized early on that having big ideas is nothing without understanding the business behind making them work. As a sophomore, he enrolled in classes at the MIT Sloan School of Management, where many of his classmates had the opposite problem: “A lot of people there have really great ideas, but less technical talent,” he says.

This realization led to Trotman’s first small business, Gogeshe, which was essentially a small team of technology consultants who worked on design, development, and marketing. Clients could come in with a business model for an undeveloped product, and Trotman’s group would flesh out the idea into a full-scale design and prototype. Clients could then either keep the Gogeshe team on to ultimately develop the product, or buy out the design and take it elsewhere for development.

Trotman’s mind is restless, though, and almost as soon as Gogeshe was off the ground, he was already focusing on a new idea: making use of latent processing power in mobile phones. At the end of a summer internship at Intel Corp., he won a hackathon with this idea, which was only 24 hours old at that point. Trotman realized that with more than 24 hours, there could be immense potential in fleshing out the project into another small business.

Along with partner Alex Koren, an undergraduate at Johns Hopkins University, Trotman spent a year contacting investors, entering competitions, and developing Hyv, a company that facilitates the borrowing of unused processing power from phones, and lending it to companies that need extra power.

“It’s sort of philanthropic,” Trotman says. “While you’re playing Angry Birds, you could have a tiny slice of data analytics running in the background of your phone, potentially helping someone do their cancer research, for example.”

Unfortunately, technological limits made the idea hard to scale: Phones are susceptible to slow networks and WiFi connections, which made it hard for Hyv to compete with running the same type of business through computers. For now, Hyv is on the back burner — but Trotman is confident that once phone technology gets faster, his business will have a place.

In the near future, however, he plans to stick with his interest in eliminating inefficiencies: After graduating from MIT this June, Trotman will begin a job at Bloomberg, the New York software, data, and media company, where he will work to streamline the process of investing.

HOW AN LED-FILLED “ROBOT GARDEN” CAN MAKE CODING MORE ACCESSIBLE

Read more in the February 26, 2015 CSAIL feature by Adam Conner-Simon, CSAIL Communications, also posted below and appearing in the MIT News.  [Image above: A distributed robot garden system created by students in CSAIL to teach younger students about algorithms and programming. Photo/Jason Dorfman, CSAIL]

Here’s one way to get kids excited about programming: a "robot garden" with dozens of fast-changing LED lights and more than 100 origami robots that can crawl, swim, and blossom like flowers.

A team from CSAIL and the Department of Mechanical Engineering have developed a tablet-operated system that illustrates their cutting-edge research on distributed algorithms via robotic sheep, origami flowers that can open and change colors, and robotic ducks that fold into shape by being heated in an oven.

In a paper recently accepted to the 2015 International Conference on Robotics and Automation, researchers describe the system’s dual functions as a visual embodiment of their latest work in distributed computing, as well as an aesthetically appealing way to get more young students, and particularly girls, interested in programming.

The system can be managed via tablet or any Bluetooth-enabled device, either through a simple “control by click” feature that involves clicking on individual flowers, as well as a more advanced “control by code” feature where users can add their own commands and execute sequences in real-time.

“Students can see their commands running in a physical environment, which tangibly links their coding efforts to the real world,” says Lindsay Sanneman, who is lead author on the new paper. “It’s meant to be a launchpad for schools to demonstrate basic concepts about algorithms and programming.”

Each of the system’s 16 tiles are connected via Arduino microcontrollers and programmed via search algorithms that explore the space in different ways, including a “graph-coloring” algorithm that ensures that no two adjacent tiles ever share the same color.

“The garden tests distributed algorithms for over 100 distinct robots, which gives us a very large-scale platform for experimentation,” says CSAIL Director Daniela Rus, the Andrew and Erna Viterbi Professor of Electrical Engineering and Computer Science and a co-author of the paper. “At the same time, we hope that it also helps introduce students to topics like graph theory and networking in a way that’s both beautiful and engaging.”

Rus previously helped develop a distributed system of robots that watered, harvested, and took various metrics of an actual vegetable garden using complex motion-planning and object-recognition algorithms.

Among the other researchers on the new project were PhD candidate Joseph DelPreto, postdocs Ankur Mehta and Shuhei Miyashita, and members of MIT Professor Sangbae Kim’s Biomimetics Robotics Lab, including undergraduates Deborah Ajilo and Negin Abdolrahim Poorheravi, among others.

Kim’s team developed eight distinct varieties of origami flowers — including lilies, tulips, and birds of paradise — which are embedded with printable motors that he says “allow them to blossom in very interesting ways.” The sheep robots were created via traditional print-and-fold origami techniques, while the magnet-powered ducks started as two-dimensional paper prints that were heated in an oven, causing them to automatically fold into shape.

“Many elements of the garden can be made very quickly, including the pouch motors and the LED flowers,” DelPreto says. “We’re hoping that rapid fabrication techniques will continue to improve to the point that something like this could be easily built in a standard classroom.”


Sanneman and DelPreto showed off the current garden to local schools at CSAIL’s “Hour of Code” event in December, and say that they plan to incorporate it into a programming curriculum involving printable robots that they have developed for middle and high schools.

In the future, they also hope to make the garden operable by multiple devices simultaneously, and may even experiment with interactive auditory components by adding microphones and music that would sync to movements.

Members of the Optical and Quantum Communications Group in the Research Lab of Electronics (RLE) including Prof. Jeffrey Shapiro with senior research scientist Franco Wong and postdoc Zheshen Zhang have shown in a recent paper in Physical Review Letters that entanglement can improve optical sensor performance even if doesn't survive light's interaction with the environment. The work could make possible the development of non-invasive techniques of quantum sensing with potential applications in biomedicine.

Read more in the March 9, 2015 MIT News Office article by Larry Hardesty titled "Quantum sensor’s advantages survive entanglement breakdown - Preserving the fragile quantum property known as entanglement isn’t necessary to reap benefits," also posted below.

The extraordinary promise of quantum information processing — solving problems that classical computers can’t, perfectly secure communication — depends on a phenomenon called “entanglement,” in which the physical states of different quantum particles become interrelated. But entanglement is very fragile, and the difficulty of preserving it is a major obstacle to developing practical quantum information systems.

In a series of papers since 2008, members of the Optical and Quantum Communications Group at MIT’s Research Laboratory of Electronics have argued that optical systems that use entangled light can outperform classical optical systems — even when the entanglement breaks down.

Two years ago, they showed that systems that begin with entangled light could offer much more efficient means of securing optical communications. And now, in a paper appearing in Physical Review Letters, they demonstrate that entanglement can also improve the performance of optical sensors, even when it doesn’t survive light’s interaction with the environment.

“That is something that has been missing in the understanding that a lot of people have in this field,” says senior research scientist Franco Wong, one of the paper’s co-authors and, together with Jeffrey Shapiro, the Julius A. Stratton Professor of Electrical Engineering, co-director of the Optical and Quantum Communications Group. “They feel that if unavoidable loss and noise make the light being measured look completely classical, then there’s no benefit to starting out with something quantum. Because how can it help? And what this experiment shows is that yes, it can still help.”

Phased in

Entanglement means that the physical state of one particle constrains the possible states of another. Electrons, for instance, have a property called spin, which describes their magnetic orientation. If two electrons are orbiting an atom’s nucleus at the same distance, they must have opposite spins. This spin entanglement can persist even if the electrons leave the atom’s orbit, but interactions with the environment break it down quickly.

In the MIT researchers’ system, two beams of light are entangled, and one of them is stored locally — racing through an optical fiber — while the other is projected into the environment. When light from the projected beam — the “probe” — is reflected back, it carries information about the objects it has encountered. But this light is also corrupted by the environmental influences that engineers call “noise.” Recombining it with the locally stored beam helps suppress the noise, recovering the information.

The local beam is useful for noise suppression because its phase is correlated with that of the probe. If you think of light as a wave, with regular crests and troughs, two beams are in phase if their crests and troughs coincide. If the crests of one are aligned with the troughs of the other, their phases are anti-correlated.

But light can also be thought of as consisting of particles, or photons. And at the particle level, phase is a murkier concept.

“Classically, you can prepare beams that are completely opposite in phase, but this is only a valid concept on average,” says Zheshen Zhang, a postdoc in the Optical and Quantum Communications Group and first author on the new paper. “On average, they’re opposite in phase, but quantum mechanics does not allow you to precisely measure the phase of each individual photon.”

Improving the odds

Instead, quantum mechanics interprets phase statistically. Given particular measurements of two photons, from two separate beams of light, there’s some probability that the phases of the beams are correlated. The more photons you measure, the greater your certainty that the beams are either correlated or not. With entangled beams, that certainty increases much more rapidly than it does with classical beams.

When a probe beam interacts with the environment, the noise it accumulates also increases the uncertainty of the ensuing phase measurements. But that’s as true of classical beams as it is of entangled beams. Because entangled beams start out with stronger correlations, even when noise causes them to fall back within classical limits, they still fare better than classical beams do under the same circumstances.

“Going out to the target and reflecting and then coming back from the target attenuates the correlation between the probe and the reference beam by the same factor, regardless of whether you started out at the quantum limit or started out at the classical limit,” Shapiro says. “If you started with the quantum case that’s so many times bigger than the classical case, that relative advantage stays the same, even as both beams become classical due to the loss and the noise.”

In experiments that compared optical systems that used entangled light and classical light, the researchers found that the entangled-light systems increased the signal-to-noise ratio — a measure of how much information can be recaptured from the reflected probe — by 20 percent. That accorded very well with their theoretical predictions.

But the theory also predicts that improvements in the quality of the optical equipment used in the experiment could double or perhaps even quadruple the signal-to-noise ratio. Since detection error declines exponentially with the signal-to-noise ratio, that could translate to a million-fold increase in sensitivity.

“This is a breakthrough,” says Stefano Pirandola, an associate professor of computer science at the University of York in England. “One of the main technical challenges was the experimental realization of a practical receiver for quantum illumination. Shapiro and Wong experimentally implemented a quantum receiver, which is not optimal but is still able to prove the quantum illumination advantage. In particular, they were able to overcome the major problem associated with the loss in the optical storage of the idler beam.”

“This research can potentially lead to the development of a quantum LIDAR which is able to spot almost-invisible objects in a very noisy background,” he adds. “The working mechanism of quantum illumination could in fact be exploited at short-distances as well, for instance to develop non-invasive techniques of quantum sensing with potential applications in biomedicine.”

MIT announced a major thrust toward addressing cybersecurity with the launch of three new initiatives including one focused on technology research to be based in the Computer Science and Artificial Intelligence Lab (CSAIL). Read more.

Read more in the March 13, 2015 MIT News Office article by Larry Hardesty titled "MIT launches three new cybersecurity initiatives - Complementary programs will address cybersecurity from the perspectives of technology, public policy, and organizational management," also posted below.

Computer-network security breaches are never out of the news for long, but lately, they’ve been hogging the headlines: the Sony hack, the Uber hack, and last month, the revelation that an international gang of cybercriminals had used malware to steal an estimated billion dollars from financial institutions over two years.

In this context, MIT yesterday announced plans to address the problem of cybersecurity from three angles: technology, public policy, and organizational management.

At an event at MIT’s Stata Center, the home of the Computer Science and Artificial Intelligence Laboratory (CSAIL), with more than 200 students, academics, and industry representatives in attendance, MIT faculty and administrators unveiled three new cybersecurity initiatives, to be housed at CSAIL and the MIT Sloan School of Management. Funded with a $15 million grant from the Hewlett Foundation, the MIT Cybersecurity Policy Initiative will pool the expertise of researchers at CSAIL, MIT Sloan, the MIT departments of political science and economics, and the Science, Technology, and Society program to better characterize the security dynamics of large networked systems, with the aim of guiding policymakers.

Cybersecurity@CSAIL will provide funding and coordination for the lab’s ongoing research into hardware- and software-based approaches to computer security, while MIT Sloan’s Interdisciplinary Consortium for Improving Critical Infrastructure in Cybersecurity, or (IC)3, will focus on the human element — how organizations can ensure that their employees or volunteers are not creating security vulnerabilities, whether intentionally or not.

The launch event was emceed by Maria Zuber, MIT’s vice president for research, and the speakers included representatives of each of the three initiatives and MIT President L. Rafael Reif.

In his opening remarks, Reif emphasized both the new initiatives’ partnerships with industry and the interdependence of their research programs. “New technologies will require new policies and incentives,” he said. “Emerging policies must adapt to future technologies. And none of that matters if they cannot make the present a safe place to do business.”

“Security by default”

Reif was followed by Daniela Rus, the Viterbi Professor of Electrical Engineering and Computer Science and director of CSAIL. Rus began by emphasizing MIT’s long history of involvement in cybersecurity: as the home of one of the first computers to allow multiple simultaneous users, it was also the birthplace of the computer password.

But Rus also gave some sense of what the future of cybersecurity would look like. “Many of today’s cybersecurity issues stem from older, poorly designed systems that viewed security as an afterthought,” she said. “Organizations learned to ‘patch and pray,’ planning to manage attacks as they happened rather than fighting them systematically. But we can change that. Instead of looking for patches, we can move towards security by default.”

Rus then introduced Howard Shrobe, a principal research scientist at CSAIL, who will direct Cybersecurity@CSAIL. Shrobe elaborated on Rus’s historical observations, pointing out that the researchers who developed MIT’s multiuser computer, under the auspices of Project MAC, in fact wrote an operating system that had “security by default.” But the computers of the time simply weren’t powerful enough to execute its security protocols efficiently.

Today, however, “on every criterion that you can think of, machines are 50,000 times more powerful than when Project MAC started,” Shrobe said. “We can now start to use those resources to enforce security in a systematic way.”

Cybersecurity@CSAIL, Shrobe added, would focus on three themes: prevention, or designing systems that are harder to hack; resilience, or designing systems that can offer secure transactions even after they’ve been compromised; and regeneration, or designing systems that can repair themselves when breaches are detected.

The founding member companies of Cybersecurity@CSAIL are BAE Systems, BBVA, Boeing, BP and Raytheon.

Square one

Danny Weitzner, a CSAIL principal research scientist and director of the new Cybersecurity Policy Initiative, took the podium next. No one, Weitzner said — neither researchers nor policy makers — has a very good understanding of the dynamics of cybersecurity. But that doesn’t prevent policy makers from trying to control them. “The United States government, in last year’s budget, is spending over $13 billion on cybersecurity efforts,” Weitzner said. “That’s 1 percent of discretionary spending.” Weitzner then offered an example from his own two-year stint as the U.S. deputy chief technology officer for Internet policy. During that time, he said, the U.S. Congress was debating the Stop Online Piracy Act, which included what Weitzner called a “seemingly simple proposal to require Internet service providers to use some features of the domain name system to block access to [content pirates’] websites.”

Discussions of the proposal elicited a letter from 83 distinguished Internet engineers — including MIT’s David Clark, who was the Internet’s chief architect for most of the 1980s — who argued that tampering with the Internet’s domain name system, which translates human-readable URLs into machine-readable IP addresses, could have potentially disastrous consequences. “Their intuition as really good Internet engineers was that it could cause some problems,” Weitzner said. “But really, there was no science presented, no formal model of the interaction between the domain name system and the rest of the Internet — certainly no understanding of how individuals would behave at large scales.” Generating that type of multidisciplinary model is one of the goals of the Cybersecurity Policy Initiative.

Human factors

S. P. Kothari, the Gordon Y. Billard Professor in Management and deputy dean at MIT Sloan, then introduced the final speaker, Stuart Madnick, the Maguire Professor of Information Technologies at MIT Sloan and a professor of engineering systems, who will lead (IC)3. “It’s great to hear about the work being done to improve the technology by our colleagues at CSAIL and the regulatory considerations being studied by CPI,” Madnick said. “But various studies have shown that up to 80 percent of the incidents [of cybersecurity breaches] are aided or abetted by authorized users.” “Understanding the organizational, managerial, and strategic issues about cybersecurity is of great importance to protecting our critical infrastructure,” he added, “and that is the focus of (IC)3.”

Five faculty are promoted to full professor; three faculty are promoted to associate professor

MIT Electrical Engineering and Computer Science Department Head Anantha Chandrakasan and Associate Department Heads Bill Freeman, Silvio Micali, and David Perreault announced in February 2015, the promotions of eight faculty members in the department. Professors Adalsteinsson, Daniel, Golland, and Torralba are promoted to full professor. Professors Chilpala, Polyanskiy and Vaikuntanathan are promoted to associate professor. The promotions are effective July 1, 2015.

Elfar Adalsteinsson is a world leader in the development and application of magnetic resonance imaging (MRI). The primary goal of his research is to advance the physics and engineering of MRI to develop new methods and instruments that can be deployed in the clinic. A remarkable feature of his work is that it spans the spectrum from studies of the basic physics of MRI and the development of enabling technologies to testing and application of new imaging modalities in the clinic. For example, he has been a leader in both the development of parallel RF transmission to realize improved MRI scanning and in development of methods for measuring parameters associated with oxygenation in the brain. He is also an excellent teacher, and has been responsible for both developing an MRI course for graduate students and senior undergraduates, 6.556/HST.580, and – along with his colleagues – co-lead development of the new undergraduate course “Introduction to EECS from a Medical Perspective (6.S02)”.

Luca Daniel works on computational techniques for modeling and design of complex systems, including for microsystems (e.g., integrated circuit modeling and design) and biomedical applications (e.g., electromagnetic analysis for magnetic resonance imaging). His research encompasses the development of computationally-efficient integral equations solvers (e.g., “field solvers”), parameterized model order reduction techniques and methods for uncertainty quantification, and their embodiment in useful software tools. His work has been widely recognized, most recently in a 2014 best paper award from the IEEE Transactions on Computer Aided Design, and is having tremendous practical impact in applications such as rapid electromagnetic field prediction for MRI scanners. He has also made key educational contributions, including heavily updating and expanding the content of MIT’s flagship class on numerical simulation, 6.336J and working with colleagues to revamp the department's undergraduate header course 6.013 (“Electromagnetics and Applications”).

Polina Golland studies the shape and functions of biological structures through the statistical analysis of biomedical images. She builds computational models of the anatomical and functional variability within populations, and develops methods to detect and characterize changes in those distributions under the influence of development or disease. Her models give insight into the functional organization of the brain and into the causes of its variability. Her group releases open-source software packages for wide impact and dissemination. Prof. Golland has played a major role in developing three important classes in our EECS curriculum--two very popular graduate classes on inference and information (6.437, 6.438), and the department’s new undergraduate class on inference (6.008).

Jing Kong is an expert on the synthesis of low-dimensionality (1-D and 2-D) materials using chemical vapor deposition (CVD). For example, her widely-cited work at MIT on CVD growth of single- and few-layer graphene films is considered foundational, and has led to the ability to grow large-area high-quality graphene films and transfer them onto arbitrary substrates, assisting the explosive growth in the field. She is making similarly important contributions to CVD synthesis of few-layer hexagonal Boron Nitride and transition-metal dichalcogenides such as MoS2. Moreover, through extensive collaborations she is having substantial impact on the engineering application of these new materials in many kinds of systems. Prof. Kong is also a highly dedicated teacher who is liked by students and colleagues alike, and has contributed to the new undergraduate / graduate class “Introduction to Nanoelectronics”, (6.096/6.975) and created a new graduate-level class “Science, Technology and Applications of Carbon Nanoelectronics” (6.976).

Antonio Torralba has received wide recognition for pointing out the importance of context for object recognition — that the objects near another object help us to recognize the object itself — and developing techniques to exploit context. In collaboration with Aude Oliva, they introduced "scene recognition" as an area of study within computer vision, and developed a representation designed to capture such contextual information. Prof. Torralba developed several widely-used datasets that help to advance the field. Since tenure, he has continued with very strong research contributions, deepening his work on scene understanding, analyzing image features, studying database bias, and exploring computational photography. His teaching is strong, and he has contributed new courses within his research area, and to the deparment's large Introduction to EECS class, 6.01.

Adam Chlipala's research addresses the software development process, applying formal logic to prove programs are correct using a computer proof assistant. He aims to reduce the human cost of program verification so that it may one day become a standard part of software development. This research could ultimately increase the reliability and security of software. He received an NSF CAREER award, and he has written a book that is highly regarded and widely used within the program verification community. He has developed a new class on interactive theorem proving.

Yury Polyanskiy works in the area of information theory, including topics such as finite blocklength coding, strong data-processing inequalities, combinatorial and geometrical aspects of Hamming spaces. His pioneering work on channel dispersion, which captures the variation of the realized data rate over a channel, has been particularly impactful, and has opened up a new field of research in information theory. Prof. Polyanskiy is the winner of several awards including the prestigious Information Theory best paper award and the NSF CAREER Award. In addition to excellent undergraduate teaching, Prof. Polyanskiy has made contributions to both Information Transmission (6.441) and Fundamentals of Probability (6.436).

Vinod Vaikuntanathan studies cryptography, a topic of ever-increasing importance in modern society. He has made breakthroughs that bring us much closer to being able to compute on encrypted data, important for secure cloud computing, as well as in functional cryptography, the ability to share only some parts of an encrypted system. He has won numerous awards, including a Sloan Research Fellowship, an NSF CAREER award, and a Microsoft Faculty Fellowship. He is a gifted teacher, and he has a strong record of service within his research community.

EECS alumna Cynthia Breazeal, SM '93, ScD '00, associate professor of media arts and sciences at MIT, and members of her group in the Media Lab, the Personal Robotics Group, has developed a system that allows pre-school children to program interactive robots to serve as a introduction to programming principles as well as a research tool ultimately leading to better integration of interactive robots in educational curricula.

Read more in the March 11, 2015 MIT News Office article by Larry Hardesty titled "Teaching programming to preschoolers - System that lets children program a robot using stickers embodies new theories about programming languages," also posted below.

Researchers at the MIT Media Laboratory are developing a system that enables young children to program interactive robots by affixing stickers to laminated sheets of paper.

Not only could the system introduce children to programming principles, but it could also serve as a research tool, to help determine which computational concepts children can grasp at what ages, and how interactive robots can best be integrated into educational curricula.

Last week, at the Association for Computing Machinery and Institute of Electrical and Electronics Engineers’ International Conference on Human-Robot Interaction, the researchers presented the results of an initial study of the system, which investigated its use by children ages 4 to 8.

“We did not want to put this in the digital world but rather in the tangible world,” says Michal Gordon, a postdoc in media arts and sciences and lead author on the new paper. “It’s a sandbox for exploring computational concepts, but it’s a sandbox that comes to the children’s world.”

In their study, the MIT researchers used an interactive robot called Dragonbot, developed by the Personal Robots Group at the Media Lab, which is led by associate professor of media arts and sciences Cynthia Breazeal. Dragonbot has audio and visual sensors, a speech synthesizer, a range of expressive gestures, and a video screen for a face that can assume a variety of expressions. The programs that children created dictated how Dragonbot would react to stimuli.

“It’s programming in the context of relational interactions with the robot,” says Edith Ackermann, a developmental psychologist and visiting professor in the Personal Robots Group, who with Gordon and Breazeal is a co-author on the new paper. “This is what children do — they’re learning about social relations. So taking this expression of computational principles to the social world is very appropriate.”

Lessons that stick

The root components of the programming system are triangular and circular stickers — which represent stimuli and responses, respectively — and arrow stickers, which represent relationships between them. Children can first create computational “templates” by affixing triangles, circles, and arrows to sheets of laminated paper. They then fill in the details with stickers that represent particular stimuli — like thumbs up or down — and responses — like the narrowing or widening of Dragonbot’s eyes. There are also blank stickers on which older children can write their own verbal cues and responses.

Researchers in the Personal Robotics Group are developing a computer vision system that will enable children to convey new programs to Dragonbot simply by holding pages of stickers up to its camera. But for the purposes of the new study, the system’s performance had to be perfectly reliable, so one of the researchers would manually enter the stimulus-and-response sequences devised by the children, using a tablet computer with a touch-screen interface that featured icons depicting all the available options.

To introduce a new subject to the system, the researchers would ask him or her to issue an individual command, by attaching a single response sticker to a small laminated sheet. When presented with the sheet, Dragonbot would execute the command. But when it’s presented with a program, it instead nods its head and says, “I’ve got it.” Thereafter, it will execute the specified chain of responses whenever it receives the corresponding stimulus.

Even the youngest subjects were able to distinguish between individual commands and programs, and interviews after their sessions suggested that they understood that programs, unlike commands, modified the internal state of the robot. The researchers plan additional studies to determine the extent of their understanding.

Paradigm shift

The sticker system is, in fact, designed to encourage a new way of thinking about programming, one that may be more consistent with how computation is done in the 21st century.

“The systems we’re programming today are not sequential, as they were 20 or 30 years back,” Gordon says. “A system has many inputs coming in, complex state, and many outputs.” A cellphone, for instance, might be monitoring incoming transmissions over both Wi-Fi and the cellular network while playing back a video, transmitting the audio over Bluetooth, and running a timer that’s set to go off when the rice on the stove has finished cooking.

As a graduate student in computer science at the Weizmann Institute of Science in Israel, Gordon explains, she worked with her advisor, David Harel, on a new programming paradigm called scenario-based programming. “The idea is to describe your code in little scenarios, and the engine in the back connects them,” she explains. “You could think of it as rules, with triggers and actions.” Gordon and her colleagues’ new system could be used to introduce children to the principles of conventional, sequential programming. But it’s well adapted to scenario-based programming.

“It’s actually how we think about how programs are written before we try to integrate it into a whole programming artifact,” she says. “So I was thinking, ‘Why not try it earlier?’”

The Optical Society (OSA) announced on March 1, 2015, the selection of Professor James G. Fujimoto as the recipient of the Frederic Ives Medal / Quinn Prize. He is recognized for pioneering the field of optical coherence tomography (OCT) and for his research group’s and collaborator’s contributions leading to its widespread medical application and major commercial impact.

The Frederic Ives Medal is the highest award of the OSA and recognizes overall distinction in optics. It is awarded at the plenary of the OSA's Annual Meeting. Other MIT faculty who have received the Ives Medal include Professor Erich P. Ippen, for laying the foundations of ultrafast science, and the late Professor Hermann A. Haus, for fundamental and seminal contributions to the understanding of quantum noise in optical systems.

Prof. Fujimoto studied at MIT (SB ’79, SM ’81, and PhD ’84), completing his doctoral research under the supervision of Prof. Erich Ippen in ultrafast optics. Now Elihu Thomson Professor of Electrical Engineering, Fujimoto joined the faculty in the MIT Electrical Engineering and Computer Science Department in 1985. As a principal investigator in the Research Laboratory of Electronics (RLE), Fujimoto’s group and clinical collaborators are credited with the invention and development of optical coherence tomography (OCT). Their landmark publication in in Science in 1991 has remained one of the highest cited papers in biomedical optics and ushered in a new era in clinical biomedical optical imaging.

Working with Mr. Eric Swanson at MIT, Fujimoto co-founded a start-up company that was acquired by Carl Zeiss and lead to the development of OCT in ophthalmology. OCT is now a standard imaging modality in ophthalmology for the detection and treatment monitoring of macular degeneration, diabetic retinopathy and glaucoma. An estimated 20-30 million ophthalmic imaging procedures are performed worldwide each year. Mr. Swanson and Fujimoto also co-founded a second MIT start-up that developed OCT for intravascular imaging, where it is an emerging technology for assessing atherosclerotic plaque and percutaneous therapy such as stenting. Hundreds of researchers worldwide work on OCT in diverse fields such as cardiology, endoscopy and cancer surgery. Last year the global sales of OCT systems exceeded $400 million produced by more than 36 OCT systems companies.

Fujimoto has been influential as an educator, training numerous researchers who have become leaders in the fields of photonics and biomedicine. He has served as co-chair of international meetings such as the Conference on Lasers and Electro Optics, the European Conferences on Biomedical Optics, and Ultrafast Phenomena. Since 2003, he has served as co-chair of the SPIE Biomedical Optics symposium, the largest international meeting on biophotonics. From 2000 to 2003 Fujimoto served as Director of the Optical Society of America and is currently a director of the International Society for Optics and Photonics (SPIE). This year, Fujimoto was also awarded an honorary doctorate from the Nicolaus Copernicus University in Torun Poland.

Read more:

James Fujimoto Awarded the Honorary Doctorate Degree at the Nicolaus Copernicus University

Biomedical Optical Imaging and Biophotonics Group

“So you are in! [as a postdoctoral associate in EECS at MIT] What is important about your research that matters in applying for a job? Is there funding? Is your [research] area one that has traction in getting funding?”

Professor Munther Dahleh, spoke extemporaneously to an overflow crowd of over 100 postdocs who came to the first fall event held in late October 2014 for Postdoc6, a group, which was created by the Electrical Engineering and Computer Science (EECS) Department in late 2013. Dahleh, the Director of MIT’s Engineering Systems Division and the William A. Coolidge Professor in electrical engineering and computer science is talking with the group to encourage open discussion about the issues that these researchers face as they seek to establish their paths in either academic research or research positions in industry. He guided the responsive group towards his main point: “Ownership — own your piece of work. Show that you can think for yourself.”

Dilip Krishnan, a former postdoc since fall 2013, worked with Professor Bill Freeman in the field of computer vision to develop depth perception. When he graduated from NYU, he didn’t know whether he wanted to head for an academic or industry research position, so building his research independence as a postdoc was his most important next step. “In a PhD,” he said, “things are driven by the vision of your advisor rather than your own. A postdoc gives you much more freedom to set your own agenda but you need to take advantage of that opportunity,” Krishnan notes. He is now a research scientist at Google Cambridge.

At MIT, as of fall 2014 there are 1,565 postdoctoral associates and in the EECS Department roughly 240 postdocs working in the four affiliate labs — Computer Science and Artificial Intelligence (CSAIL), Laboratory for Information and Decision Systems (LIDS), Microsystems Technology Laboratories (MTL) and Research Laboratory of Electronics (RLE). Besides being spread across these four labs housed in buildings 39, 38, 36 and 32, the population of EECS postdocs pales in comparison with roughly 800 undergraduates and 750 graduate students. The ultimate effect is that EECS postdocs, who are already steeped in heavy demands for original research and leadership in their fields — usually in a short time span — are likely to become isolated, missing potential opportunities for networking that could make a difference as they reach for their next, or ultimate career goal.

Recognizing these realities and at the suggestion of the 2013 EECS Visiting Committee, the leadership in EECS decided to start a new group — now called Postdoc6 — to give its postdoctoral associates a sense of community and to address concerns and needs. After an initial gathering in late fall 2013 to test the interest, EECS department head Anantha Chandrakasan and then associate head Bill Freeman launched the group with a daylong workshop “A Bootcamp for EECS Postdocs.” The attendance was strong and feedback clearly showed that the variety of talks and participatory aspects of the class were useful and appreciated.

Freeman continued to establish the new EECS postdoc group organizing three other events in spring 2014 to answer the needs typical for postdocs: networking, understanding funding avenues, starting a teaching or industry career and learning from recent graduates who were interviewing for faculty positions. He noted: “Being a post-doc is the ideal job. You usually have the freedom to explore whatever you want to, and you're often in a very supportive work environment. You generally have the freedom to try something new, and the freedom to fail without bad consequences happening.”

Time to gain perspective

Former CSAIL postdoc I-Ting Angelina Lee, who earned SM and PhD degrees in computer science from MIT, found her experience valuable in the long run, though she says that the transitional (and temporary) nature of the position created anxiety for her. Deciding between an industry position and an academic one was difficult. Having taught 6.172 alongside Prof. Charles Leiserson, Lee found that she gained perspective on the role of a principal investigator (pi) and professor. “ I got to see different perspectives from the point of view of a pi but not total head of household— if the house comes crashing down, Charles would be there,” she noted last summer (2014). She is now an assistant professor at Washington University in St. Louis.

Networking with other postdocs and graduate students is very helpful

Gunjan Agarwal a former postdoc in Prof. Dana Weinstein’s group in the Microsystems Technology Laboratories, MTL, was very enthusiastic about Postdoc6 and hoped for even more events and opportunities for postdocs to interact. As a Course 2, MIT MechE PhD graduate, Agarwal noted about postdocs in general, “Interaction with other postdocs is important — you are on a different curve than a graduate student. As a postdoc, different possibilities open up — it's a big choice. While I was a grad student, I would interact with other students even though we were not doing the same thing — on a broad perspective.” While still a postdoc Agarwal noted that just as EECS graduate students meet for coffee hours, it would be beneficial for a similar, regular event that could be organized by and for postdocs in the department. She also suggested establishing an online portal site for postdocs to be privately listed with the potential to exchange resource ideas and encourage greater mutual support.

Balancing act: the two-body problem

When Daniel Zoran graduated from Hebrew University and joined Prof. Freeman’s lab to work in computer vision, he had been well prepared as a graduate student for his postdoc position and preferred to work at MIT. “I came here to expand my horizons. I hope to become faculty at some point. I chose MIT because it’s very densely populated in a good way.”

Zoran’s wife, a neuro biologist was also looking for a postdoc position. They were unusually fortunate to be able to choose between three offers. His wife took a postdoc position in MIT’s Brain and Cognitive Sciences and expects to stay on for three years. “My wife does real experiments where everything is very time-consuming,” he notes. “With a two-year-old son, very little free time is left. The bottom line is that you need to publish something during this span of time.” Zoran plans to continue research in Cambridge while his wife finishes another year of her postdoc at MIT.

Postdoc positions in Engineering — early preparation

Ram Vasudevan, former postdoc with Prof. Russ Tedrake in CSAIL and now an assistant professor at the University of Michigan, came to MIT after earning his PhD at UC Berkeley. After attending several of the Postdoc6 events in spring 2014, he wishes he had had this kind of preparation while he was a graduate student. In reflection, he says “I like how you guys are now turning it [Postdoc6] into something that is not only for postdocs but for senior graduate students. That is a smart idea. Many places are just beginning to start up similar initiatives in engineering now.”

Having searched for a long time for the most advantageous postdoc position, Vasudevan found that the academic landscape in engineering — in terms of sequence of study followed by postdoc positions — is beginning to resemble that in biology. “That is,” he notes, “the postdoc is starting to become a standard component of any individual interested in pursuing an academic career.”

Broad vision in a defined field

Zheshan Zhang has been a postdoc for three years — working as an experimentalist in quantum optics in RLE. Working with theorist Prof. Jeff Shapiro and experimentalist Franco Wang is an experience that Zhang both enjoys and finds very helpful in his personal development. Quantum optics as a field is in its early stages, but Zhang is hopeful for opportunities as some companies such as Google and IBM have begun to set up labs in this developing field.

Raised in China, where he completed his undergraduate work, Zhang was a graduate student at Georgia Tech in both the US and Georgia Tech’s European campus in France. He notes, “I have lived in 3 different continents. This is a benefit.” With multinational contacts, Zhang hopes to continue to give research talks abroad to make his work known — while he explores both academia and industry.

As the EECS department leadership has shaped the structuring of Postdoc6 to respond to feedback from its postdocs, several participants have voiced appreciation and suggestions for building what is already a responsive community. In addition to making data on postdoc positions available to its current graduate students, EECS postdocs have voiced the need for establishing a database on the job scene — for both academic and industry positions. Several have suggested a career fair for postdocs and the need for local opportunities for coffee hours and networking.

Postdoc 6 — full cycle

By January 2015 launching into its second year, Postdoc6 held another all-day workshop. “With many new postdocs every year, we should allow for this repeat,” said the new coordinator for Postdoc6 Aude Oliva, Principal Research Scientist in CSAIL and the MIT Computer Vision and Graphics Group. “Now we have the full cycle,” she notes, “… and continuity is underway.”

Oliva should know about postdocs. She held four different postdoc positions in four different countries, in four different research areas over the course of seven years. “Those are the golden years of building your mind!” she notes enthusiastically about the postdoc experience. In fact, Oliva, formerly an associate professor in the MIT Department of Brain and Cognitive Sciences, used to run workshops for graduate students about postdocs and future advising. “I am a big fan of the future,” she says. “When you are neither a student or a faculty,” she notes, “you can open your mind up and dedicate most of your time to develop a research program that will make you unique.”

Now almost two years in CSAIL and EECS, Oliva finds that the cross-disciplinary nature of these communities allows for the best opportunities for exchange “…where one of my students in neuroscience, for example, can talk with an expert in robotics,” she says. She also notes that larger talks and events offered by Big Data and Start6 are also geared to postdocs.

Based on the attendance (over 70) at the January 26 workshop, Oliva says that the feedback suggested concrete views, such as how the search process happens behind the scenes, and personal perspectives such as developing an effective research statement, were elements most highly appreciated. Professor Charles Leiserson led a session for participants to develop a unique research statement in two sentences. “Learning how to make such an impactful statement was so useful!” Oliva said. She was glad to have been present herself.

Recognized for pioneering contributions to high-performance THz quantum-cascade lasers

Qing Hu, the MIT Distinguished Professor in Electrical Engineering and Computer Science and Principal Investigator in the Research Laboratory of Electronics (RLE) has been selected by the Board of Directors of the Optical Society (OSA) as the 2015 recipient of the Nick Holonyak, Jr. Award. He is recognized for his pioneering contributions to high-performance THz quantum-cascade lasers and their applications in imaging and sensing.

The Nick Holonyak Jr. Award was established in 1997 in honor of Nick Holonyak, Jr., who has made distinguished contributions to the field of optics through the development of semiconductor-based light emitting diodes and semiconductor lasers.

Professor Hu has made significant contributions to physics and device applications over a broad electromagnetic spectrum, from millimeter wave, through terahertz (THz), to infrared frequencies. His research has involved technology development for detectors and sources, as well as system-level imaging and sensing applications. Among those contributions, the most distinctive is his development of high-performance THz quantum cascade lasers. This breakthrough has already found applications in sensing and real‐time THz imaging, which was also pioneered by his group, the Millimeter-wave and Terahertz Devices Group. His work, reported in the journal Nature Physics, on achieving sold-state terahertz lasers operable at closer to room temperature was recognized in late 2010 for bringing the possibility of making solid-state lasers as a promising means of detecting trace explosives much closer to reality.

Professor Hu is a Fellow of the Optical Society of America (OSA), of the American Physical Society (APS), of the IEEE, and of the American Association for the Advancement of Science (AAAS). He is the recipient of the 2012 IEEE Photonics Society William Streifer Scientific Achievement Award. He has been an Associate Editor of Applied Physics Letters since 2006, and was the co-chair of the 2006 International Workshop on Quantum Cascade Lasers.

In addition to his research, Professor Hu has also made important contributions to the department in service and teaching. He has served on the EECS faculty search committee during 2008-2011, the EECS ABET committee during 2012-2013, and the personnel committee since 2012. He has taught a broad range of courses, including signals and systems (6.003), microelectronic devices and circuits (6.012), electromagnetics (6.013/6.014), quantum mechanics (6.017 prior to 1995), and solid-state physics (6.730 and 6.732).

Computer Science and Artificial Intelligence Laboratory (CSAIL) principal investigator and EECS Prof. Martin Rinard with members of his research group, the Center for Resilient Software, including CSAIL research scientist Stelios Sidiroglou-Douskos have developed DIODE (for Directed Integer Overflow Detection) a system to provide an effective mechanism for finding dangerous integer overflows that affect memory allocation sites in debugging code.  The group will present a new algorithm for identifying integer-overflow bugs at this month's Association for Computing Machinery's (ACM's) International Conference on Architecural Support for Programming Languages and Operating Systems.  

Read more in the MIT News Office March 24, 2015 article by Larry Hardesty titled "Better debugger - System to automatically find a common type of programming bug significantly outperforms its predecessors," also posted below.

Integer overflows are one of the most common bugs in computer programs — not only causing programs to crash but, even worse, potentially offering points of attack for malicious hackers. Computer scientists have devised a battery of techniques to identify them, but all have drawbacks.

This month, at the Association for Computing Machinery’s International Conference on Architectural Support for Programming Languages and Operating Systems, researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) will present a new algorithm for identifying integer-overflow bugs. The researchers tested the algorithm on five common open-source programs, in which previous analyses had found three bugs. The new algorithm found all three known bugs — and 11 new ones.

The variables used by computer programs come in a few standard types, such as floating-point numbers, which can contain decimals; characters, like the letters of this sentence; or integers, which are whole numbers. Every time the program creates a new variable, it assigns it a fixed amount of space in memory.

If a program tries to store too large a number at a memory address reserved for an integer, the operating system will simply lop off the bits that don’t fit. “It’s like a car odometer,” says Stelios Sidiroglou-Douskos, a research scientist at CSAIL and first author on the new paper. “You go over a certain number of miles, you go back to zero.”

In itself, an integer overflow won’t crash a program; in fact, many programmers use integer overflows to perform certain types of computations more efficiently. But if a program tries to do something with an integer that has overflowed, havoc can ensue. Say, for instance, that the integer represents the number of pixels in an image the program is processing. If the program allocates memory to store the image, but its estimate of the image’s size is off by several orders of magnitude, the program will crash.

Charting a course

Any program can be represented as a flow chart — or, more technically, a graph, with boxes that represent operations connected by line segments that represent the flow of data between operations. Any given program input will trace a single route through the graph. Prior techniques for finding integer-overflow bugs would start at the top of the graph and begin working through it, operation by operation.

For even a moderately complex program, however, that graph is enormous; exhaustive exploration of the entire thing would be prohibitively time-consuming. “What this means is that you can find a lot of errors in the early input-processing code,” says Martin Rinard, an MIT professor of computer science and engineering and a co-author on the new paper. “But you haven’t gotten past that part of the code before the whole thing poops out. And then there are all these errors deep in the program, and how do you find them?”

Rinard, Sidiroglou-Douskos, and several other members of Rinard’s group — researchers Eric Lahtinen and Paolo Piselli and graduate students Fan Long, Doekhwan Kim, and Nathan Rittenhouse — take a different approach. Their system, dubbed DIODE (for Directed Integer Overflow Detection), begins by feeding the program a single sample input. As that input is processed, however — as it traces a path through the graph — the system records each of the operations performed on it by adding new terms to what’s known as a “symbolic expression.”

“These symbolic expressions are complicated like crazy,” Rinard explains. “They’re bubbling up through the very lowest levels of the system into the program. This 32-bit integer has been built up of all these complicated bit-level operations that the lower-level parts of your system do to take this out of your input file and construct those integers for you. So if you look at them, they’re pages long.”

Trigger warning

When the program reaches a point at which an integer is involved in a potentially dangerous operation — like a memory allocation — DIODE records the current state of the symbolic expression. The initial test input won’t trigger an overflow, but DIODE can analyze the symbolic expression to calculate an input that will.

The process still isn’t over, however: Well-written programs frequently include input checks specifically designed to prevent problems like integer overflows, and the new input, unlike the initial input, might fail those checks. So DIODE seeds the program with its new input, and if it fails such a check, it imposes a new constraint on the symbolic expression and computes a new overflow-triggering input. This process continues until the system either finds an input that can pass the checks but still trigger an overflow, or it concludes that triggering an overflow is impossible.

If DIODE does find a trigger value, it reports it, providing developers with a valuable debugging tool. Indeed, since DIODE doesn’t require access to a program’s source code but works on its “binary” — the executable version of the program — a program’s users could run it and then send developers the trigger inputs as graphic evidence that they may have missed security vulnerabilities.

“DIODE provides an effective mechanism for finding dangerous integer overflows that affect memory allocation sites, the source of many critical security vulnerabilities,” says Cristian Cadar, a senior lecturer in computing at Imperial College London. “DIODE is based on symbolic execution, a state-of-the-art technique that provides the ability to automatically explore and analyze paths through a program by modeling these paths as mathematical formulas. In DIODE, symbolic execution is specifically optimized to find integer overflows that affect memory allocation sites, by enhancing it with a novel exploration mechanism that enables it to synthesize dangerous inputs that reach the overflow target. On the practical side, DIODE operates directly on binaries, making it easy to find critical bugs and security vulnerabilities.”

Former and current EECS graduate students have created new methods to automate identification of potential areas for development in rural villages in both India and sub-Saharan Africa. The group won a $10,000 prize last year from the MIT IDEAS Global Challenge. [Image above: To test their software system for identifying houses in satellite images, the first step was to manually determine the house locations. In this sample satellite image of a rural village in India (top), the team created a map (bottom), where red indicates ‘‘building,’’ white indicates ‘‘not building,’’ and green indicates ‘‘not sure,’’ which includes pixels very close to the building boundaries. Courtesy of the researchers]

Read more in the March 23, 2015 MIT News Office article by David Chandler titled "Satellite imagery can aid development projects. New image-analysis methods can automate identification of cost-effective sites for grants or microgrids," also posted below.

Projects that target aid toward villages and rural areas in the developing world often face time-consuming challenges, even at the most basic level of figuring out where the most appropriate sites are for pilot programs or deployment of new systems such as solar-power for regions that have no access to electricity. Often, even the sizes and locations of villages are poorly mapped, so time-consuming field studies are needed to locate suitable sites.

Now, a team of graduate students at MIT and a social-service group of data scientists have come up with a way of automating parts of that evaluation process, by developing software that can identify houses and even types of houses from readily-available satellite imagery — potentially saving considerable time that would otherwise be spent sending teams from village to village. Their findings have now been published in the journal Big Data.

The multidisciplinary team came together in the course of discussions at MIT’s Sidney Pacific graduate dormitory, explains team member Brian Spatocco: “We started talking about this problem, and we realized we all had skills that were relevant.” The team’s proposal gained them a $10,000 prize last year from the MIT IDEAS Global Challenge, which helped get the project rolling and enabled team members to visit rural areas in India last summer to test their image-processing system against conditions on the ground. The group, which grew to include an MIT alumnus at New York-based DataKind DataCorps and another researcher there, focused on two initial projects, in India and Africa, though they stress that their software solutions could be applied to many other kinds of projects and other regions.

Selecting villages for aid

The first project was to select villages in sub-Saharan Africa for a program of unrestricted cash grants to help people in low-income rural areas improve their standard of living by enabling them to buy equipment, livestock, or whatever they felt best met their needs. The system adopted by the grant-giving agency was to target the poorest villages, selected by counting the percentage of houses with thatched roofs compared with those topped by more expensive metal roofs — a task that had been carried out by fieldworkers on the ground.

The second project was selecting villages in rural parts of India for installation of microgrids to supply electricity from solar panels and battery-storage systems, and then figuring out the optimum sites for those panels and the most efficient network configuration for distributing that power.

In both cases, the key first element is automating the task of figuring out where the buildings are within a satellite image. For this research, the team used two kinds of satellite imagery: Google Earth, which has three color “channels” in their images, corresponding to red, green, and blue, and commercial satellite imagery that also includes a near-infrared channel that provides additional information for detecting vegetation and other features.

Identifying structures

The process begins by having people examine the satellite images visually and pick out the houses. These manually-selected examples are then entered in as training data for a machine-learning system that attempts to generalize the criteria for determining what is a house and what isn’t, and then “can try to predict, in a new image,” where the houses are, says George Chen, a co-author of the paper. One of the challenges, he explains, is that it’s not always clear whether a given structure is two houses close together, or two parts of the same building. In other cases, “the house color is similar to the ground color,” though that’s less common, he says.

But as more examples get processed by the system, “over time, the computers can learn from the hand-picked set” and get better at figuring out where the houses are, Spatocco says. Then, in the case of the African aid project, which is currently making unconditional cash transfers in villages in Kenya and Uganda, an additional step is used to distinguish houses that have metal roofs, which are much more reflective than thatched ones.

In the project for installing microgrids in India, once the locations of houses are determined, the computer runs thousands of different variations of where solar panels, battery packs, and distribution wires could be located. This allows the team to pick the configurations that can provide power to the greatest number of houses with the least wiring needed, to minimize the costs. The program can also select configurations based on other local criteria, such as a village that specifically wants its solar panels in a particular location.

The team says that the general algorithms they’ve developed could have many other uses beyond the two specific projects they initially tested. For example, there is little data on demographic changes in India, in terms of which areas have gained or lost population and by how much, and Spatocco says “this could be an extremely powerful tool” for analyzing those population shifts by automated tracking of where houses are and how that changes over time. “It could answer deep questions about these demographic dynamics,” he adds.

As the project continues, four villages will be selected in India for the next phase of testing: Two will have solar microgrids installed using existing methods, and two will have them installed using the patterns selected by the software. These villages, selected to be as closely matched as possible, will then be compared over time for the actual costs and performance of the systems, to determine exactly how much benefit can be gained from the new approach.

“We're hoping that public agencies eventually see the wisdom of mapping 100 million rural households in developing countries,” says Stewart Craine, chair of the UN Foundation’s mapping group and head of DevelopmentMaps.org, a company that offers satellite-based mapping services for development organizations, but was not associated with this project. “Preliminary mapping can reduce wasting expensive field-time mapping households, and spend more on village discussions and fine-tuning of the preliminary desktop design,” he explains. Overall, he says, this paper is “an excellent contribution” to the field.

The team also included MIT graduate students Kendall Nowocin, Vivek Sakhrani, and Ling Xu, as well as Kush Varshney SM ’06, PhD ’10, and Brian Abelson, both of Datakind DataCorps. The team worked with Bangalore-based nonprofit SELCO Foundation, which is carrying out the microgrid installations in India, and with GiveDirectly for the cash transfer program in Africa. The research received mentorship and support from the Tata Center for Technology and Design at MIT, which is part of the MIT Energy Initiative.

by Adam Conner-Simons, CSAIL

Michael Stonebraker, a researcher at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) who has revolutionized the field of database management systems (DBMSs) and founded multiple successful database companies, has won the Association for Computing Machinery’s (ACM) A.M. Turing Award, often referred to as “the Nobel Prize of computing.” This year marks the first time that the Turing Award comes with a Google-funded $1 million prize.

In its announcement today, ACM said that Stonebraker “is being recognized for fundamental contributions to the concepts and practices underlying modern database systems. Stonebreaker is the inventor of many concepts that were crucial to making databases a reality and that are used in almost all modern database systems. His work on INGRES introduced the notion of query modification, used for integrity constraints and views. His later work on Postgres introduced the object-relational model, effectively merging databases with abstract data types while keeping the database separate from the programming language." (See the ACM Turing Award announcement.)

An adjunct professor of computer science and engineering at MIT and a principal investigator at CSAIL, Stonebraker sometimes jokes that he didn’t know what he was researching for more than 30 years. “But then, out of nowhere, some marketing guys started talking about ‘big data,’” he says. “That’s when I realized that I’d been studying this thing for the better part of my academic life.”

Stonebraker's work over the past four decades has helped spur a multibillion-dollar “big data” industry that he himself has participated in, creating and leading nine separate companies, including VoltDB, Tamr, Paradigm4, and Vertica (which was bought by Hewlett-Packard in 2011 for $340 million).

“Mike has been a trailblazer in the field of databases by asking the essential questions about how we collect, organize, and access information in our lives,” says Daniela Rus, director of CSAIL and the Andrew and Erna Viterbi Professor of Electrical Engineering and Computer Science. “He has been both a devoted academic and a serial entrepreneur, and all of us at CSAIL are so inspired by his work and proud to have him as a colleague.”

In his previous work at the University of California at Berkeley, Stonebraker developed two of his most influential systems, Ingres and Postgres, which provide the foundational ideas — and, in many cases, specific source code — that spawned several contemporary database products, including IBM’s Informix and EMC’s Greenplum.

Ingres was one of the first relational databases, which provide a more organized way to store multiple kinds of entities – and which now serve as the industry standard for business storage. Larry Rowe, a professor emeritus at Berkeley who helped Stonebraker commercialize the technology, remembers that many of their colleagues didn’t think that relational databases could evolve from academic theory to practical application.

“There’s no better way to get Mike going than to tell him, ‘You can’t do that,’” Rowe says. “One of his great abilities is to imagine something that should exist, ask himself why it doesn’t, and then set his new research goal to be making it happen.”

Postgres, meanwhile, integrated Ingres’ ideas with object-oriented programming, enabling users to natively map objects and their attributes into databases. This new notion of “object-relational” databases could be used to represent and manipulate complex data, like computer-aided design, geospatial data, and time series.

Stonebraker’s major projects at MIT include:

• C-Store, a column-oriented database that partitions tables by column, allowing delivery of dramatic performance speedups for reading large quantities of data;
• H-Store, a parallel database management system that can deliver a high sustained rate of operations (“transactions”) per second; and
• SciDB, which represents data as arrays and provides substantially improved performance for many modern analytics settings, such as machine learning and statistical data processing.

Notably, in an era in which the term “open source” didn’t yet exist, Stonebraker also released many of his systems into the public domain, ensuring their widespread adoption and allowing other academics to build on his work.

“I am delighted that Professor Stonebraker has been recognized for his groundbreaking contributions to database technology,” says Anantha Chandrakasan, the Joseph F. and Nancy P. Keithley Professor in Electrical Engineering, and head of MIT’s Department of Electrical Engineering and Computer Science (EECS). “This award is an honor for the EECS department and inspiring for the many members of the MIT community who have worked with him.”

Stonebraker co-directs CSAIL’s Intel Science and Technology Center for Big Data with Sam Madden, a professor of computer science and engineering. Before joining MIT, Stonebraker was a professor of computer science at Berkeley for 29 years. A graduate of Princeton University, he earned his master's degree and his PhD from the University of Michigan.

“It’s every computer scientist’s dream to get this award, and I am so very honored to be selected,” Stonebraker says. “It reinforces and validates the importance of the work that I have been doing alongside so many other researchers in the field of database management systems.”

Past Turing Award recipients who have either taught at or earned degrees from MIT include Shafi Goldwasser and Silvio Micali (2013), Barbara Liskov (2008), Ronald Rivest (2002), Manuel Blum (1995), Butler Lampson (1992), Fernando Corbato (1990), Ivan Sutherland (1988), John McCarthy (1971) and Marvin Minsky (1969).

Stonebraker will formally receive the award during the ACM’s annual awards banquet on June 20 in San Francisco.

In a broad new assessment of the status and prospects of solar photovoltaic technology, MIT researchers including Vladimir Bulović, Associate Dean for Innovation and the Fariborz Maseeh (1990) Professor of Emerging Technology and Joel Jean, EECS graduate student and lead author in the journal Energy & Environmental Science say that it is “one of the few renewable, low-carbon resources with both the scalability and the technological maturity to meet ever-growing global demand for electricity.”

[Image: Illustration shows the MIT team’s proposed scheme for comparing different photovoltaic materials, based on the complexity of their basic molecular structure. The complexity increases from the simplest material, pure silicon (single atom, lower left), to the most complex material currently being studied for potential solar cells, quantum dots (molecular structure at top right). Materials shown in between include gallium aresenide, perovskite and dye-sensitized solar cells. Courtesy of the researchers.]

Read more in the March 26, 2015 MIT News Office article by David L. Chandler titled "Analysis sees many promising pathways for solar photovoltaic power - New study identifies the promise and challenges facing large-scale deployment of solar photovoltaics," also posted below.

Use of solar photovoltaics has been growing at a phenomenal rate: Worldwide installed capacity has seen sustained growth averaging 43 percent per year since 2000. To evaluate the prospects for sustaining such growth, the MIT researchers look at possible constraints on materials availability, and propose a system for evaluating the many competing approaches to improved solar-cell performance.

The analysis is presented in the journal Energy & Environmental Science; a broader analysis of solar technology, economics, and policy will be incorporated in a forthcoming assessment of the future of solar energy by the MIT Energy Initiative.

The team comprised MIT professors Vladimir Bulović, Tonio Buonassisi, and Robert Jaffe, and graduate students Joel Jean and Patrick Brown. One useful factor in making meaningful comparisons among new photovoltaic technologies, they conclude, is the complexity of the light-absorbing material.

The report divides the many technologies under development into three broad classes: wafer-based cells, which include traditional crystalline silicon, as well as alternatives such as gallium arsenide; commercial thin-film cells, including cadmium telluride and amorphous silicon; and emerging thin-film technologies, which include perovskites, organic materials, dye-sensitized solar cells, and quantum dots.

With the recent evolution of solar technology, says Jean, the paper’s lead author, it’s important to have a uniform framework for assessment. It may be time, he says, to re-examine the traditional classification of these technologies, generally into three areas: silicon wafer-based cells, thin-film cells, and “exotic” technologies with high theoretical efficiencies.

“We’d like to build on the conventional framework,” says Jean, a doctoral student in MIT’s Department of Electrical Engineering and Computer Science. “We’re seeking a more consistent way to think about the wide range of current photovoltaic technologies and to evaluate them for potential applications. In this study, we chose to evaluate all relevant technologies based on their material complexity.”

Under this scheme, traditional silicon — a single-element crystalline material — is the simplest material. While crystalline silicon is a mature technology with advantages including high efficiency, proven reliability, and no material scarcity constraints, it also has inherent limitations: Silicon is not especially efficient at absorbing light, and solar panels based on silicon cells tend to be rigid and heavy. At the other end of the spectrum are perovskites, organics, and colloidal quantum dots, which are “highly complex materials, but can be much simpler to process,” Jean says.

The authors make clear that their definition of material complexity as a key parameter for comparison does not imply any equivalency with complexity of manufacturing. On the contrary, while silicon is the simplest solar-cell material, silicon wafer and cell production is complex and expensive, requiring extraordinary purity and high temperatures.

By contrast, while some complex nanomaterials involve intricate molecular structures, such materials can be deposited quickly and at low temperatures onto flexible substrates. Nanomaterial-based cells could even be transparent to visible light, which could open up new applications and enable seamless integration into windows and other surfaces. The authors caution, however, that the conversion efficiency and long-term stability of these complex emerging technologies is still relatively low. As they write in the paper: “The road to broad acceptance of these new technologies in conventional solar markets is inevitably long, although the unique qualities of these evolving solar technologies — lightweight, paper-thin, transparent — could open entirely new markets, accelerating their adoption.”

The study does caution that the large-scale deployment of some of today’s thin-film technologies, such as cadmium telluride and copper indium gallium diselenide, may be severely constrained by the amount of rare materials that they require. The study highlights the need for novel thin-film technologies that are based on Earth-abundant materials.

The study identifies three themes for future research and development. The first is increasing the power-conversion efficiency of emerging photovoltaic technologies and commercial modules.

A second research theme is reducing the amount of material needed per cell. Thinner, more flexible films and substrates could reduce cell weight and cost, potentially opening the door to new approaches to photovoltaic module design.

A third important research theme is reducing the complexity and cost of manufacturing. Here the researchers emphasize the importance of eliminating expensive, high-temperature processing, and encouraging the adoption of roll-to-roll coating processes for rapid, large-scale manufacturing of emerging thin-film technologies.

“We’ve looked at a number of key metrics for different applications,” Jean says. “We don’t want to rule out any of the technologies,” he says — but by providing a unified framework for comparison, he says, the researchers hope to make it easier for people to make decisions about the best technologies for a given application. Martin Green, a professor at the Australian Centre for Advanced Photovoltaics at the University of New South Wales who was not involved in this work, says the MIT team has produced “some interesting new insights and observations.” He says the paper’s main significance “lies in the attempt to take a unifying look at the issues involved in choosing between PV technologies.”

“The issues involved are complex,” Green adds, “and the authors abstain from betting on any particular PV technology.”

Students and graduates of Prof. Rob Miller's group, the User Interface Design Group have designed a system for visualizing and exploring thousands of solutions to a programming problem, ultimately enhancing online teaching and learning. Members of the group including first author and EECS graduate student Elena Glassman will present their work in April at the Association for Computing Machinery's Conference on Human Factors in Computing Systems. [Photo above: MIT graduate students Elena Glassman and Jeremy Scott discuss their work on OverCode. Photo by: Jose-Luis Olivares/MIT]

Read more in the March 30, 2015 MIT News Office article by Larry Hardesty titled "Reviewing online homework at scale - System clusters similar student programs together, so instructors can identify broad trends," also posted below.

In computer-science classes, homework assignments consist of writing programs. It’s easy to create automated tests that determine whether a given program yields the right outputs to a series of inputs. But those tests say nothing about whether the program code is clear or confusing, whether it includes unnecessary computation, and whether it meets the terms of the assignment. Professors and teaching assistants review students’ code to try to flag obvious mistakes, but even in undergraduate lecture courses, they usually don’t have time for exhaustive analysis. And that problem is much worse in online courses, with thousands of students, each of whom might have approached a problem in a slightly different way.

In April, at the Association for Computing Machinery’s Conference on Human Factors in Computing Systems, MIT researchers will present a new system that automatically compares students’ solutions to programming assignments, lumping together those that use the same techniques.

For each approach, the system — called OverCode — creates a program template, using variable names that a preponderance of students happen to have converged on. It then displays templates side-by-side, graying out the code they share, so the differences stand out in relief. And from any template, instructors can, if they choose, pull up a list of the student programs that accord with it.

Instructors who notice variations across templates that make no difference in practice can also write rules establishing the equivalence of alternatives. In some instances, for example, “y*x” might yield a different result than “x*y”, but — depending on the ways in which x and y are defined — in other instances, it won’t. When it doesn’t, an instructor could further winnow down the number of templates by creating the rule “y*x = x*y”.

The system could allow instructors of online courses to provide generalized feedback that addresses a broader swath of their students. But it could also provide information on how computer-science courses — both online and on campus — could be better designed.

With online courses, “in a few months, you can have many orders of magnitude of students go through the same material and find all the interesting alternative solutions or make the same errors,” says Elena Glassman, an MIT graduate student in computer science and engineering and first author on the new paper. “Then it’s taking all those records of what people did and making sense of it so that when we run the course again, it’s better, and when we run the course residentially, we’re better able to handle the particular 200 students that we’re meeting with on a regular basis.”

Two programs that perform the same computation may have code that looks somewhat different. The programmers may have chosen different variable names — “total,” say, in one case, versus “result” in the other. Subfunctions may be executed in different orders.

So in addition to comparing programs’ code, OverCode observes the values that variables take on as the programs execute. Two programs with variables that take on the same values in the same order are judged to be identical.

In their new paper, Glassman and her collaborators — her thesis advisor, professor of computer science and engineering Rob Miller; her fellow graduate student Jeremy Scott; Rishabh Singh, who completed his PhD at MIT last year and is now at Microsoft Research; and Philip Guo, an assistant professor of computer science at the University of Rochester — also report the results of two usability studies that evaluated OverCode.

In the studies, 24 experienced programmers reviewed thousands of students’ solutions to three introductory programming assignments, using both OverCode and a standard tool that displays solutions one at a time. For each assignment, the subjects were given 15 minutes to assess the strategies students most commonly used to design a particular function and to provide general feedback on each, complete with example code.

Remarkably, when assessing the simplest of the three assignments, the subjects analyzing raw code performed as well those using OverCode: In both cases, the five strategies they identified covered about half of the student responses.

For the most difficult of the three assignments, however, the OverCode users covered about 45 percent of student responses, while the subjects analyzing raw data covered only about 9 percent. “The strategy starts to shine on more-complicated programs,” Glassman says.

MIT is one of three schools to receive an Alfred P. Sloan Foundation grant to create a “University Center of Exemplary Mentoring” (UCEM) that will focus on the recruitment, retention, and academic success of underrepresented minority doctoral students. Four departments in the MIT School of Engineering (SoE) including the Electrical Engineering and Computer Science Department (EECS) and Biological Engineering (BE), Chemical Engineering (ChemE) and Mechanical Engineering (MechE) will be part of this three-year program that will recruit 36 underrepresented doctoral students. Leslie A. Kolodziejski, professor and graduate officer of electrical engineering and computer science will serve as a member of the  governing body for the UCEM. [Photo: Christine Ortiz (top left), dean for graduate education at MIT, is the principal investigator of a new Alfred P. Sloan Foundation University Center of Exemplary Mentoring (UCEM) grant. Leading the UCEM will be (clockwise from top center) Leslie Kolodziejski of the Department of Electrical Engineering and Computer Science, David Hardt of the Department of Mechanical Engineering, Eboney Hearn of the Office of the Dean for Graduate Education, Forest White of the Department of Biological Engineering, and Patrick Doyle of the Department of Chemical Engineering.]

Read more in the April 2, 2015 MIT News Office article by David R. Gordon, Office of Foundation Relations titled "Alfred P. Sloan Foundation awards grant to create a University Center of Exemplary Mentoring at MIT," also posted below.

MIT has received a major grant — one of only three awarded nationally — from the Alfred P. Sloan Foundation to create a “University Center of Exemplary Mentoring” (UCEM) that will focus on the recruitment, retention, and academic success of underrepresented minority doctoral students in four departments in the School of Engineering: Biological Engineering, Electrical Engineering and Computer Science, Chemical Engineering and Mechanical Engineering. See the Sloan Foundation article about the new Exemplary Mentoring initiative. 

The MIT UCEM will carry out strategic and customized recruitment strategies tailored to the four participating departments, as well as a structured program that addresses key barriers to retention and educational success consisting of academic support, mentoring, and personal and professional development opportunities.

Under the terms of a three-year, $840,000 grant, 36 underrepresented doctoral students will be recruited to participate in the UCEM.

MIT Dean for Graduate Education and the Morris Cohen Professor of Materials Science and Engineering Christine Ortiz, who is the principal investigator overseeing the initiative, noted that the grant will benefit individual students, the participating departments, and the Institute as a whole. “Aside from the significant positive impact on the participating students,” she said, “the UCEM will greatly enrich MIT’s diversity efforts overall. It will enhance synergy and momentum, awareness, and sharing of best practices within the four participating departments. It also will make possible the expansion and development of programmatic and assessment methods in the recruitment and retention of underrepresented doctoral students.” She added that each of the participating departments is known for its strong record of success in recruiting underrepresented minority doctoral students and is eager to be part of the UCEM’s programming.

The UCEM will be an attractive incentive for prospective students thinking about matriculating to MIT. “We also plan to use part of the Sloan funding to provide all of the scholars with a variety of networking and community building activities,” Ortiz added. “We see this grant as a direct way to promote the diversity of our student body.”

Blanche E. Staton, senior associate dean for graduate education, noted that the grant from the Sloan Foundation would enable MIT “to build on our long track record of success in encouraging underrepresented students to matriculate in doctoral programs in the STEM fields.” She added that the review committee made a special point of praising efforts such as the MIT Summer Research Program, begun in 1986, which has brought talented undergraduates to spend the summer preparing for graduate work, and the CONVERGE workshop, an intensive four-day preparatory weekend aimed at successfully introducing new students to graduate study at MIT.

Staton also cited the inherent importance of diversity in “helping us tackle and solve world problems. A more diverse student body, by definition, will strengthen our entire community by bringing diverse perspectives and talent to bear on our common challenges.”

When MIT initially received an invitation from the Sloan Foundation to participate in this competition, President L. Rafael Reif welcomed the opportunity to address the challenge directly. “MIT is deeply committed to recruiting and ensuring the success of graduate students from underrepresented minority groups,” he wrote to the foundation. “The opportunity to participate in the Sloan Foundation’s Minority PhD Program would enable us to enhance our existing efforts, which are critical to increasing the pipeline of minority PhD graduates in STEM fields.”

A team of faculty from the four participating departments — Leslie A. Kolodziejski, professor and graduate officer of electrical engineering and computer science; David Hardt, professor and graduate officer of mechanical engineering; Patrick Doyle, associate professor and graduate officer of chemical engineering; and Forest White, professor and graduate officer of biological engineering — as well as Eboney Hearn, assistant dean for diversity initiatives in the Office of the Dean for Graduate Education (ODGE) will serve as the governing body for the UCEM. In addition, the team will leverage Institute-wide efforts focused on community, equity, inclusion, and diversity through physics professor and Institute Community and Equity Officer Edmund Bertschinger, as well as through Blanche Staton.

“This has been a wonderful team effort,” said Ortiz in reflecting on the development and production of a winning proposal. “President Reif and his senior leadership team have all been very supportive. The Office of Foundation Relations and the Office of Institutional Research have been tremendously helpful resources. And members of my staff, along with the dean, faculty, and staff in the School of Engineering, have come together to create a great program.”

“In bringing the four departments together under the ‘umbrella’ of a UCEM,” Ortiz continued, “we expect to enhance and synergize our efforts to create, share, and implement effective practices in recruitment, retention, climate, and academic success for URM students. In turn, we will capitalize on our UCEM and adapt approaches that can be applied Institute-wide to the benefit of all graduate students.”

The Sloan Foundation initially invited approximately 30 leading universities to participate in the UCEM competition. More details can be found on the Sloan Foundation website.

The Society for Industrial and Applied Mathematics has named CSAIL principal investigator Charles E. Leiserson as one of its 2015 Fellows for his “enduring influence on parallel computing systems and their adoption into mainstream use through scholarly research and development.”

Leiserson introduced the concept of cache-oblivious algorithms, which are able to exploit the memory hierarchy at a near optimal level, despite not having any tuning parameters for cache size or cache-line length. He also created the Cilk multithreaded programming technology, and spurred the development of multiple Cilk-based chess-playing programs, winning numerous prizes in international competition.

A professor in MIT's Department of Electrical Engineering and Computer Science (EECS), Leiserson heads CSAIL’s Supertech Research Group, which investigates scalable computing technologies. He coauthored the influential textbook “Introduction to Algorithms”, and has developed multiple courses on algorithms and parallel programming.

This past year Leiserson was also the recipient of ACM’s Ken Kennedy Award, as well as the Paris Kanellakis Theory and Practice Award, alongside his former PhD student Robert D. Blumofe.

The Fellows program is intended to recognize excellence in research, industrial work, educational activities that reach a broad audience, or other forms of excellence directly related to the goals of SIAM.

Leiserson will be formally recognized at a special luncheon at ICIAM 2015, which will take place August 10-14 in Beijing, China.

Read more at the SIAM website: http://fellows.siam.org/index.php?sort=year&value=2015

Members of the Quantum Photonics Lab including its director Professor Dirk Englund and EECS graduate student Hannah Clevenson have developed a new, ultrasensitive magnetic-field detector that is 1,000 times more energy-efficient than its predecessors. This work, which could lead to miniaturized, battery-powered devices for medical and materials imaging, contraband detection, and geological exploration, is reported in the latest issue of Nature Physics.

[Image above: H. Clevenson/MIT Lincoln Laboratory FULL SCREEN In this image, laser light enters a synthetic diamond from a facet at its corner and bounces around inside the diamond until its energy is exhausted. This excites In this image, laser light enters a synthetic diamond from a facet at its corner and bounces around inside the diamond until its energy is exhausted. This excites "nitrogen vacancies" that can be used to measure magnetic fields. Credit: H. Clevenson/MIT Lincoln Laboratory]

Read more in the April 6, 2015 MIT News Office article by Larry Hardesty titled "Better sensors for medical imaging, contraband detection - Magnetic-field detector is 1,000 times more efficient than its predecessors," also posted below.

MIT researchers have developed a new, ultrasensitive magnetic-field detector that is 1,000 times more energy-efficient than its predecessors. It could lead to miniaturized, battery-powered devices for medical and materials imaging, contraband detection, and even geological exploration.

Magnetic-field detectors, or magnetometers, are already used for all those applications. But existing technologies have drawbacks: Some rely on gas-filled chambers; others work only in narrow frequency bands, limiting their utility.

Synthetic diamonds with nitrogen vacancies (NVs) — defects that are extremely sensitive to magnetic fields — have long held promise as the basis for efficient, portable magnetometers. A diamond chip about one-twentieth the size of a thumbnail could contain trillions of nitrogen vacancies, each capable of performing its own magnetic-field measurement.

The problem has been aggregating all those measurements. Probing a nitrogen vacancy requires zapping it with laser light, which it absorbs and re-emits. The intensity of the emitted light carries information about the vacancy’s magnetic state.

“In the past, only a small fraction of the pump light was used to excite a small fraction of the NVs,” says Dirk Englund, the Jamieson Career Development Assistant Professor in Electrical Engineering and Computer Science and one of the designers of the new device. “We make use of almost all the pump light to measure almost all of the NVs.”

The MIT researchers report their new device in the latest issue of Nature Physics. First author on the paper is Hannah Clevenson, a graduate student in electrical engineering who is advised by senior authors Englund and Danielle Braje, a physicist at MIT Lincoln Laboratory. They’re joined by Englund’s students Matthew Trusheim and Carson Teale (who’s also at Lincoln Lab) and by Tim Schröder, a postdoc in MIT’s Research Laboratory of Electronics.

Telling absence

A pure diamond is a lattice of carbon atoms, which don’t interact with magnetic fields. A nitrogen vacancy is a missing atom in the lattice, adjacent to a nitrogen atom. Electrons in the vacancy do interact with magnetic fields, which is why they’re useful for sensing.

When a light particle — a photon — strikes an electron in a nitrogen vacancy, it kicks it into a higher energy state. When the electron falls back down into its original energy state, it may release its excess energy as another photon. A magnetic field, however, can flip the electron’s magnetic orientation, or spin, increasing the difference between its two energy states. The stronger the field, the more spins it will flip, changing the brightness of the light emitted by the vacancies.

Making accurate measurements with this type of chip requires collecting as many of those photons as possible. In previous experiments, Clevenson says, researchers often excited the nitrogen vacancies by directing laser light at the surface of the chip.

“Only a small fraction of the light is absorbed,” she says. “Most of it just goes straight through the diamond. We gain an enormous advantage by adding this prism facet to the corner of the diamond and coupling the laser into the side. All of the light that we put into the diamond can be absorbed and is useful.”

Covering the bases

The researchers calculated the angle at which the laser beam should enter the crystal so that it will remain confined, bouncing off the sides — like a tireless cue ball ricocheting around a pool table — in a pattern that spans the length and breadth of the crystal before all of its energy is absorbed.

“You can get close to a meter in path length,” Englund says. “It’s as if you had a meter-long diamond sensor wrapped into a few millimeters.” As a consequence, the chip uses the pump laser’s energy 1,000 times as efficiently as its predecessors did.

Because of the geometry of the nitrogen vacancies, the re-emitted photons emerge at four distinct angles. A lens at one end of the crystal can collect 20 percent of them and focus them onto a light detector, which is enough to yield a reliable measurement.

“NV centers are very nice to work with,” says Frank Narducci, a physicist at the U.S. Naval Air Systems Command. “You just have this little solid-state sample. You don’t have to do anything to it. You don’t have to put it in a vacuum. You don’t have to cryogenically cool it. To get them excited, you can just use a green laser — a laser pointer is good enough. You don’t have to have anything super-fancy in the way of stabilized lasers.”

“What’s cool about this is that they’re using the sample itself kind of like a waveguide, to bounce the light around,” he continues. “Their sample is quite small. Because the laser doesn’t have to be anything particularly special, that could be small, too. So you could envision very small magnetometers. And correspondingly, you could make them very cheap.”

“From a Navy perspective,” he adds, “we talk about throwaway magnetometers a lot, where you might be flying over some area of the ocean and you want to make some measurements, so you just throw a handful of these out. If you get a really high-sensitivity magnetometer that’s really cheap, that would be one really good application for it.”

Munther Dahleh, the William A. Coolidge Professor in the Department of Electrical Engineering and Computer Science, will head a new Institute for Data, Systems, and Society (IDSS) to be launched on July 1. Offering a range of cross-disciplinary academic programs, including a new undergraduate minor in statistics, IDSS will be home to faculty from the Engineering Systems Division (ESD), the Laboratory for Information and Decision Systems(LIDS), and the current Sociotechnical Systems Research Center (SSRC).

Read more in the April 8, 2015 MIT News Office article titled "Deans announce new Institute for Data, Systems, and Society. MIT-wide effort aims to bring the power of data to the people," also posted below.

What do data scientists and social scientists have in common? Not nearly enough — yet. But now, MIT is creating a new institute that will bring together researchers working in the mathematical, behavioral, and empirical sciences to capitalize on their shared interest in tackling complex societal problems.

As announced today by the deans of all five of the Institute’s schools, MIT will officially launch the new Institute for Data, Systems, and Society (IDSS) on July 1. Offering a range of cross-disciplinary academic programs, including a new undergraduate minor in statistics, IDSS will be directed by Munther Dahleh, the William A. Coolidge Professor in the Department of Electrical Engineering and Computer Science.

While providing a structure and incentives for new alliances among researchers from across MIT, IDSS will become a central “home” for faculty from the Engineering Systems Division and a number of existing units, including the Laboratory for Information and Decision Systems and the Sociotechnical Systems Research Center. IDSS will also launch a new MIT center on statistics.

“The Institute for Data, Systems, and Society will be a platform for some of the most exciting research and educational activity in complex systems at MIT,” Provost Martin Schmidt says. “Its formation is the result of intensive consultations among more than three dozen faculty members over many months. Those consultations have helped define many of the challenges that need to be addressed. I am deeply grateful to Munther for his leadership throughout this process.”

“This new institute allows MIT to bring all of its strengths to bear in exciting new directions,” says Ian A. Waitz, dean of the School of Engineering. “The modern proliferation of data and networks means that every problem, solution, or idea can be modeled, tested, and analyzed in ways and on scales that were unheard of 20, or even 10, years ago. This is leading to unprecedented challenges in areas like cybersecurity, and to spectacular opportunities for innovation, as in global online learning.”

“Engineering and science are always embedded in social realities, from deeply felt cultural traditions to building codes to political tensions,” says Deborah Fitzgerald, dean of the School of Humanities, Arts, and Social Sciences. “IDSS will allow the deep, original thinking about the physical universe that is done by our scientists and engineers to come together with the rigorous work of MIT’s great social scientists and economists.”

Students and researchers working with IDSS have a “nearly infinite pool of societal challenges they can begin to address together,” Dahleh says. In fields such as energy, transportation, social networks, health care, and financial systems, the explosion of data sources and networks is redefining not only social systems and infrastructure, but many of the disciplines that investigate them.

“In order to understand things like power outages and bank failures, you still need electrical engineers and economists — but today you also need anthropologists and data scientists, too,” Dahleh says. “Our ability to collect and aggregate data is already well beyond our ability to understand what it could tell us — and no single discipline, on its own, holds the keys to solving this problem.”

New demand, new program

Dahleh is building out the IDSS leadership team, which already begun to lay the foundations of its academic and research activities. One committee created has already completed the design of an interdisciplinary undergraduate minor in statistics.

A new PhD program anchored in both analytical tools and social sciences is also in the planning stages. The PhD will be problem-driven, requiring every student to gain in-depth expertise in a wide range of analytical tools; deep understanding of a coherent program in social science; and substantial knowledge in one application domain area.

“Bringing social-science dimensions to our strengths in science and engineering will have an enormous impact,” says Michael Sipser, dean of the School of Science. “Statistics has become the fastest growing college major in the country, and the IDSS will give us an opportunity to meet this demand in a distinctively MIT way.”

Also charged with hiring new faculty, the IDSS leadership has successfully recruited a top theoretical statistician to join MIT; they will continue to identify candidates in networked systems and connection science who have broad interests in engineering, economics, and social networks. They have also organized a statistics workshop, to be held May 14 and 15, to bring together thought leaders in statistics around exciting challenges created by the new era of data-rich applications. The speakers will offer technical presentations in mathematical statistics, machine learning, econometrics, and biostatistics.