Hans Berliner was at the center of computer chess research for two decades.
Former Carnegie Mellon University School of Computer Science faculty member Hans Berliner, a world champion correspondence chess player who built the first game-playing computer ever to defeat a human champion at any game, died Jan. 13 in Riviera Beach, Fla. He was 87.
Berliner, who earned his Ph.D. in computer science at Carnegie Mellon in 1975 and served as a senior research scientist until his retirement in 1998, was at the center of computer chess research for two decades. He led the development of Hitech, the first chess computer to achieve the rank of senior master and, in 1988, the first to beat a grandmaster.
"Hans really was the father of computer chess," said Scott Fahlman, an emeritus faculty member in CMU's Language Technologies Institute (LTI). "Hans not only made major early contributions himself, but his students and those clearly influenced by his insights went on to dominate the field and to finally realize the dream of creating a chess machine that could compete with human players at the world-champion level."
Berliner was born in Berlin in 1929 and emigrated with his family in 1937 to Washington, D.C. At age 13, he learned chess from a friend one rainy day at summer camp. By age 20, he had achieved master status. He eventually became a champion in correspondence chess, in which moves are exchanged via postcard and games could last months or even years.
He won the Fifth World Correspondence Chess Championship in a match that began in 1965 and ended three years later. It was an earlier game against Yakov Estrin of Russia, however, that grandmaster Andy Soltis would rank number one in his 2000 book, "The 100 Best Chess Games of the 20th Century."
Berliner was offered a CMU job soon afterward by Herb Simon and, at age 40, he began work on computer chess and his own Ph.D.
"He was into all kinds of mental games," recalled Michael Shamos, distinguished career professor in the Institute for Software Research and LTI. One day, about 40 years ago, Shamos was playing bridge with friends in the Computer Science Department lounge in Wean Hall, while Berliner was across the room, playing chess or some other game. Berliner was listening to the bids, however, and after the first trick said, "The trump king is offside." Shamos remains amazed at Berliner's ability to determine the location of the key card in the hand based only on the bidding.
In addition to computer chess, Berliner worked on a program for the simpler game of backgammon. In 1979, the program, called BFG, beat the reigning world backgammon champion, becoming the first computer program to beat a world champion in any game.
Chess remained his focus, however. The number of possible moves in chess is astronomical, but Berliner found a way to effectively narrow the choices and make the problem computationally feasible. Called B* (or B-star), the algorithm evaluated decision trees, assigning an "optimistic" or a "pessimistic" score to each node on the tree. The idea was to prune the tree, finding a path that was sufficient to solve the problem, but not necessarily the perfect path.
This and other innovations were implemented in 1984 in Hitech. This computer used 64 very-large-scale-integrated, or VLSI, circuits, one for each square on the chessboard, which operated in parallel to help determine strategy. It could consider 175,000 positions per second.
"There was such an enthusiasm for Hitech that I've never seen before," Berliner recalled to SCS's The Link magazine in 2012. "Everyone wanted to know what the latest developments were and if they could help."
By 1987, Hitech was ranked 190th in the United States and the only computer among the top 1,000 chess players.
Shamos said Berliner had no patience for nonsense and could be curmudgeonly, "but he was completely immersed in chess."
"You can't become a top-rated chess player like Hans without being competitive and self-confident, but I never saw him as being 'over the top,'" Carl Ebeling, one of his students and the designer of Hitech's VLSI circuits, told The Link. "He led by example more than anything else. There was a constant attention to detail and he was always thinking, looking out for the next idea that might work," added Ebeling, who retired in 2012 as a computer science professor at the University of Washington.
Berliner retired to Florida in 1998. He is survived by a brother, Ernest. Read his New York Times obituary.
Carnegie Mellon University President Subra Suresh and five CMU faculty and researchers will be among global leaders at the World Economic Forum Annual Meeting in Davos-Klosters, Switzerland, Jan. 17-20.
The theme of this year's meeting is "Responsive and Responsible Leadership." The Fourth Industrial Revolution, which has been the theme in the past, continues to drive convergence of technologies that blur the lines between physical, digital and biological systems.
"Carnegie Mellon University plays a very important role in connecting technology with the human condition," Suresh said. "I'm delighted to participate with my CMU colleagues in this year's World Economic Forum. The wonderful work that takes place at CMU can thus be part and parcel of the conversation in Davos and help shape policies and practices that will emerge from these conversations."
Among this year's sessions is the "CMU Ideas Lab: Reimagining the Human Body with Carnegie Mellon University," which will take place on Tuesday, Jan. 17.
Suresh will introduce three CMU faculty members to discuss how new technologies will converge with everyday human existence. Panelists are:
- Adam Feinberg, associate professor of biomedical engineering and materials science and engineering;
- Steve Collins, associate professor of mechanical engineering and robotics; and
- Justine Cassell, professor and associate dean of Technology Strategy and Impact in the School of Computer Science.
Mariette DiChristina, editor-in-chief of Scientific American, will facilitate the session.
Suresh, who co-chairs the WEF's Future Council on Production, also will participate in a public session on the Future of Production. CMU will also be represented during the Global University Leaders Forum (GULF) annual meeting on Wednesday, Jan. 18, where Suresh will be presenting on a panel.
President and Mrs. Mary Suresh will host a reception for CMU alumni and friends at 6 p.m. Wednesday, Jan. 18, at Schneider’s Davos Café-Restaurant, Promenade 68, CH–7270 Davos Platz.
Illah Nourbakhsh, professor of robotics and director of the CREATE Lab at the Robotics Institute, and Randy Sargent, senior systems scientist and commercialization specialist, will present a continuation of their Earth Time-Lapse research with new data images focusing on urban fragility, nutritional deficiency forecasts due to climate change, violence over time, displaced person movements, tropical diseases and other data sets.
Cassell's Socially-Aware Robot Assistant (SARA), will be the only interactive demonstration in the forum's main hall.
"SARA is a virtual personal assistant who is going to help attendees find sessions that match their interests and find people who will be interesting and helpful for them to meet," Cassell said.
"The importance of the research that I'm doing is to ensure that in this increasingly technological age we maintain a focus on those aspects of humanity that are most important to us. That we ensure that we don't lose the social bonds between people and the relationships that we build with one another, which after all are what make us human," she said.
World Economic Forum sessions can be streamed online throughout the annual meeting. Join the conversation on social media using #WEF, and follow CMU faculty's experience on Twitter.
Carnegie Mellon is one of only 25 universities in the world, 12 from the U.S., that have been invited to join GULF. CMU has been a member since 2011. Business members of the forum include the top 1,000 companies from around the world. CMU also played a central role at the 2016 WEF meeting, which featured Suresh and eight faculty members.
By Julianne Mattera
The slideshow above shows a number of images by first-year art student Lumi Barron that she created of Carnegie Mellon University and Pittsburgh.
Lumi Barron likes to make art on the run.
The first-year College of Fine Arts student has been regularly painting and sketching the places she visits since she was in high school, and she continued the practice when she came to Carnegie Mellon University this fall.
Barron's drawings and watercolors take from five to 90 minutes to complete. When something catches her eye, she'll break out her sketchbook, pen and watercolors to create a new piece.
While waiting to meet with a professor one snowy morning, Barron painted the view of the campus lawn from the third floor of the CFA building. She has sketched Pittsburgh row homes in a handful of minutes while waiting for a bus. Hamerschlag Hall and a view of campus from Schenley Drive have made it into her portfolio.
Barron said the personal project is her way of recording where she has been and the experiences she has had.
"[They're] little snapshots of just that moment," she said.
Barron came to CMU's School of Art with an interest in studying animation. She was drawn in by the variety of artistic opportunities at the university, including its emphasis on the intersection between art and technology.
"When visiting CMU and seeing the work students were creating, and speaking with various people about the art and technology integration opportunities, I felt that I could go into the program interested in animation and graduate with a good degree in exactly that, but could equally well end up focusing on something entirely different and unknown," Barron said. "It was that element of flexibility in combination with cross-disciplinary opportunity, that upon seeing, I realized was exactly what I was looking for."
By Julianne Mattera
Carnegie Mellon University Provost Farnam Jahanian called for continuing investments in cybersecurity to meet the evolving challenges in securing cyberspace. He delivered his remarks during a keynote speech at the National Science Foundation's Secure and Trustworthy CyberSpace (SaTC) Principal Investigators' Meeting.
The biennial forum of the SaTC research community, held Jan. 9-11 in Arlington, Va., included top experts in academia, government and industry.
Prior to coming to CMU, Jahanian led the NSF Directorate for Computer and Information Science and Engineering, which hosted the event.
Jahanian's keynote described the co-evolution of attacks and defenses in cybersecurity. A system that was secure in the past might not be in the future and as upgrades occur, new systems introduce new vulnerabilities. As automation pervades new platforms, Jahanian cautioned that vulnerabilities will continue to threaten critical infrastructure, automotive systems, smart grids, medical devices and transportation systems.
"Cybersecurity is a multi-dimensional problem," Jahanian said. "It requires expertise from various disciplines, not just computer scientists and mathematicians, but from economists, social scientists, behavioral scientists and policymakers."
Following this holistic approach, Carnegie Mellon's CyLab brings together experts from across the entire university, spanning the fields of engineering, computer science, public policy, business and others.
Jahanian said some of the simplest security measures are not necessarily being used. A recent Duo Security Trusted Access Report estimated that 71 percent of Android mobile devices and 50 percent of iOS devices are out of date.
Future cybersecurity challenges will continue to follow internet adoption patterns and rapidly emerging technology trends.
According to Jahanian, those emerging trends include smart systems and the melding of the cyber and physical world; the explosion of data and analytics; and advances in automation and robotics.
According to research by the Software Engineering Institute (SEI) cited by Jahanian, more than 40 percent of all security issues fixed by Google in the Android platform in 2016 came from externally developed software. Thus, systems are rarely written from scratch anymore and are instead built by integrating previously developed components, often from outside organizations. While accelerating the time to market, the process also introduces security issues.
While big data has transformative implications for commerce and the economy and is increasingly critical to accelerating the pace of discovery and innovation, it also has created major security targets that motivate sophisticated hackers, including those who gained access to millions of credit and debit card accounts from Target and Home Depot.
Jahanian's address emphasized the importance of cybersecurity and data privacy as national priorities that demand sustained investment.
"Future cybersecurity challenges threaten the tightly integrated economic, political and social fabric of society," Jahanian said. "There is a need for large-scale integration, experimentation and evaluation, and continued growth in R&D investments for cybersecurity and privacy at the federal level."
Academic institutions like CMU can help to bridge the gap between research, innovation and practice.
"Institutional and academic leadership need to support faculty and researchers serving in federal agencies," Jahanian said. "This is truly a call to service for the community."
Jahanian was one of several CMU representatives who spoke during the three-day meeting.
Alessandro Acquisti, professor of information technology and public policy at CMU's H. Heinz III College and Cylab, and director of the Privacy Economics Experiments (Peex) lab, gave the meeting's final keynote address Jan. 11.
Acquisti's talk focused on the relationships between privacy, economics, and behavioral economics in a time when people disclose so much of their personal lives and data over the internet. Much of the talk revolved around two questions: Do people care about privacy, and should they?
The degree to which people care about privacy ends up depending on the context of the situation, Acquisti said. He added that both the sharing and the protection of personal data can benefit some while having a negative impact for others — a more nuanced approach than the argument that sharing data is an unalloyed economic win-win.
For instance, Acquisti said the success of platforms like Facebook makes it possible for employers to find job candidates' publicly shared personal information, such as religious affiliations or sexual preferences, which legally shouldn't be used in the hiring process. In an experiment that included four male candidates, including one whose Facebook profile represented him as Muslim and one whose Facebook profile represented him as Christian, Acquisti and his co-author (CMU's Christina Fong), found the Christian candidate to have about a 17 percent probability of being called back for an interview, whereas the Muslim candidate's probability was about 2 percent, in more conservative states in the U.S.
The appearance of having more control over their data also can lead to people having less privacy, Acquisti said. The study's findings suggested that, when users had more control over the publication of their private information, their privacy concerns decreased while their likelihood of publishing sensitive information increased.
Other CMU faculty participants were: Norman Sadeh, a professor in the School of Computer Science (SCS); Lorrie Cranor, professor in SCS and the Department of Engineering and Public Policy (EPP) who is serving as chief technologist at the U.S. Federal Trade Commission; Greg Shannon, chief scientist for the CERT Division at CMU’s Software Engineering Institute and the assistant director for cybersecurity strategy at the White House Office of Science & Technology Policy; Cleotilde Gonzalez, a research professor in the Department of Social and Decision Sciences at the Dietrich College of Humanities and Social Sciences; and Nicolas Christin, an associate research professor in SCS and EPP.
An independent institute founded by Carnegie Mellon University will receive more than $250 million to launch an advanced robotics manufacturing institute in Pittsburgh, the U.S. Department of Defense announced Friday.
The Department of Defense awarded the public-private Manufacturing USA institute to American Robotics, a nonprofit venture led by Carnegie Mellon, with more than 220 partners in industry, academia, government and the nonprofit sector nationwide. The institute will receive $80 million from the DOD, and an additional $173 million from the partner organizations. The Richard King Mellon Foundation played a particularly important role in catalyzing the CMU proposal.
The high-level award puts Pittsburgh and CMU at the center of a new wave of manufacturing, leveraging artificial intelligence, autonomy, 3-D printing and other emerging technologies to make industrial robotics more affordable for businesses of all sizes, adaptable for many uses, and able to achieve more.
Government, industry and academic leaders said this new generation of robotics has the potential to create large numbers of new jobs and fuel economic growth by putting the U.S. squarely in the lead on advanced manufacturing.
"This new institute will provide significant benefits to the region and the nation, while creating enormous opportunities for CMU scholars and researchers, and new momentum for the university," said Carnegie Mellon President Subra Suresh, who spoke at Friday's announcement at the Pentagon. "The institute, in return, will benefit from CMU's expertise in technology, as well as its strengths in policy, ethics and human interfaces that will ensure that new technologies work to benefit humankind."
Gary Fedder, CMU's vice provost for research, was one of several university faculty and officials who led the development and preparation of the proposal for the institute, and who will play a pivotal role in its establishment.
"When the DOD announced its intention to create a new institute dedicated to robotics manufacturing, we knew that CMU's historic strength in the field, leading to cutting-edge research underway today, gave us a great opportunity," said Fedder, the Howard M. Wilkoff Professor of Electrical and Computer Engineering and a professor at CMU's Robotics Institute. "We were fortunate to build a team that brought complementary strengths from across the nation."
Howie Choset, a professor in Carnegie Mellon's Robotics Institute, worked closely with Fedder in conceiving of and establishing the institute. He said the four-pronged mission of the institute is to empower American workers to compete with low-wage workers abroad; create and sustain new jobs to secure U.S. national prosperity; lower the technical, operational, and economic barriers for small- and medium- sized enterprises as well as large companies to adopt robotics technologies; and assert U.S. leadership in advanced manufacturing.
"This work has the power to benefit society broadly, and to benefit many, many potential workers and their families," Choset said.
The use of robotics is widespread in manufacturing environments but today's robots are typically expensive, singularly purposed, challenging to reprogram, and require isolation from humans for safety. Robotics are increasingly necessary for defense and other industrial manufacturing needs, but capital cost and complexity of use limit the participation of mid-size and small manufacturers. The Advanced Robotics Manufacturing Institute (ARM) will integrate the diverse collection of industry practices and institutional knowledge across many disciplines.
"The institute will tap into CMU's research strengths in collaborative robotics, sensor technologies, materials science and human-computer interaction, areas that are rapidly transforming many sectors of our economy," said CMU Provost Farnam Jahanian. "With this unique partnership between academia, government and industry, CMU is poised to bridge the gap between research, innovation and practice in the emerging field of advanced manufacturing."
ARM will achieve its mission through defense- and industry-driven, critical technology development and workforce training. ARM focuses on key industrial sectors — aerospace, automotive, electronics and textiles — defined by its partners.
ARM joins the Manufacturing USA institute network, which is a bipartisan program that brings together industry, academia and government to co-invest in the development of world-leading manufacturing technologies and capabilities. Each Manufacturing USA institute focuses on a technology area critical to future competitiveness. In addition to robotics, other areas include 3-D printing, integrated photonics or tissue fabrication. Across the Manufacturing USA institutes, the federal government has committed over $1 billion, which has been matched by over $2 billion in non-federal investment.
By Byron Spice
Play began Jan. 11 for "Brains Vs. Artificial Intelligence: Upping the Ante," a competition at Rivers Casino in Pittsburgh that pits a Carnegie Mellon University artificial intelligence called Libratus against four of the world's best professional poker players.
During the 20-day event, Libratus and the pros - Jason Les, Dong Kim, Daniel McAulay and Jimmy Chou — will play a total of 120,000 hands of Heads-Up No-Limit Texas Hold'em. The pros will split a prize purse of $200,000, while the AI and its creators — Professor Tuomas Sandholm and Ph.D. student Noam Brown - look to prove that AI can best the top high-stake players of the game.
"A lot of people throughout the AI community are watching this event carefully," said Andrew Moore, dean of Carnegie Mellon's School of Computer Science. He said beating some of the top players in the game would be a significant achievement for AI and will only be possible if Libratus can successfully bluff and otherwise mislead its human opponents.
Moore noted that the ability to outperform the best human players at an imperfect information game such as poker could lead to a host of new applications for AI. For example, he said negotiating the best price for a new car might someday be a task that people can assign to their smartphone.
But the game they are playing in Brains Vs. AI is two-player Texas Hold'em with no restrictions on bet size, an incredibly complex game that has proven elusive to solution by AI, said Sandholm, a professor of computer science. A previous CMU AI, called Claudico, was out-pointed in the first Brains Vs. AI competition in 2015. He noted that international betting sites consider Libratus a definite underdog in this contest, with odds varying between 4-to-1 and 5-to-1 against the AI.
"I'm really delighted that we got four of the top Heads-Up No-Limit Texas Hold'em specialists in the world here today," Sandholm said, making the event the ultimate test of the AI.
AIs developed by Sandholm and Brown have won the last two Annual Computer Poker Competitions, in which pokerbots play each other. Libratus represents a two-generation leap ahead from Claudico, the AI that competed in the 2015 Brains Vs. AI.
"I don't know what to expect," said Les, one of the poker pros and a veteran of the 2015 contest, comparing the new Libratus to a player who has been practicing the game in Antarctica for years and is only now beginning to play others. He said he and his fellow players consider Brains Vs. AI to be part of an important research effort.
"If we're going to test this system, we really want to push it to the absolute limit," Les said.
Sandholm noted that the Libratus AI is not specifically a poker program. Its algorithm could be applied to any number of situations that involve incomplete and misleading information, such as business negotiations, military strategy, cybersecurity and even medical treatment design.
Libratus developed its knowledge of the game and its strategy by analyzing the rules of the game, not by trying to copy the play of humans. The AI calculated its poker strategy using about 15 million core hours of computation on the Pittsburgh Supercomputing Center's Bridges computer. Ralph Roskies, scientific director of the PSC, said this use of Bridges already has generated 2 1/2 petabytes of data.
To ensure the outcome of the competition is not due to luck, the four pros will be paired to play duplicate matches - Player A in each pair will receive the same cards as the computer receives against Player B, and vice versa. One of the players in each of these pairs will play on the floor of the casino, while his counterpart will be isolated in a separate room.
For this second installment of Brains Vs. AI, the pros have agreed to increase the number of hands to improve the chance of reaching statistical significance, that is, ruling out with high confidence the possibility that either the humans or the computer win by just getting lucky. To do so, the pros will play more days and will "two-table," playing two hands simultaneously.
Play will begin at 11 a.m. each day at Rivers Casino and end around 7 p.m. The public is welcome to observe game play, which will be in Rivers' Poker Room. Libratus' games also will be streamed via Twitch. Aggregate scores will be posted each evening on the competition website.
Brains Vs. AI is sponsored by GreatPoint Ventures, Avenue4Analytics, TNG Technology Consulting GmbH, the journal Artificial Intelligence, Intel and Optimized Markets, Inc. Carnegie Mellon's School of Computer Science has partnered with Rivers Casino, the Pittsburgh Supercomputing Center (PSC) through a peer-reviewed XSEDE allocation, and Sandholm's Electronic Marketplaces Laboratory for this event.
By Byron Spice
Computer scientists at Carnegie Mellon University have spent years inventing and perfecting a platform that uses workstations, distributed computers, mobile devices or embedded devices to solve large machine learning problems efficiently and effectively. Now they've spun off a company, Petuum Inc., to make those capabilities available commercially.
Eric Xing, professor of computer science, founder and CEO of the company, said the company has obtained $15 million in venture capital funding and expects to have the first products in the market early this year.
Machine learning (ML) and artificial intelligence (AI) technologies are key to innovations such as self-driving cars, speech recognition, computer vision, natural language processing and analysis of electronic medical records, and many other enterprise big data analysis applications.
"In 10 to 20 years, AI and ML will be the dominant workload of any computing device," Xing said. "We need to optimize how AI/ML programs are designed, programmed and run on such devices, especially as these programs grow in size and sophistication. In many areas, such as self-driving cars, current limits on most AI/ML solutions — often hand-crafted black boxes — have become a bottleneck."
"Petuum promises to be a transformative platform, enabling AI/ML programs to be built easily and to be mounted and run on different hardware platforms, using standardized methods that are transparent and repeatable. The technology in the Petuum platform allows the programs to run correctly, quickly, at scale and using minimal computing resources," Xing said.
"Our platform will make disparate computing devices, from data centers to mobile and embedded platforms, look and function like a single computer," he said.
Xing noted the massive data sets required for many large AI/ML problems already exist across these devices and Petuum will allow the AI/ML programs to operate seamlessly together. Petuum seeks to enhance and expand the use of artificial intelligence and machine learning at the much larger scales possible with distributed computing. Communication between computing devices can be tricky for AI/ML, Xing acknowledged, but he and his Sailing Lab team with collaborators have developed parameter servers, managed communication and load-balancing methods over the last several years that automatically keep the devices running in synchrony.
Though other groups have solved machine learning problems using distributed devices, Xing and his team have shown their approach provides an optimal, efficient solution for all types of ML problems, not just certain subsets such as deep learning. The platform thus can support a wide range of applications, such as natural language processing, image and video understanding, and anomaly detection in transaction data.
"We reached a point where we couldn't go further without capital investment," Xing said, prompting the launch of the company.
He expects to hire 30 to 50 people in the next six months and, because of the need for highly trained computer scientists and engineers, intends to keep the company in proximity to Carnegie Mellon.
"Our goal is to build in Pittsburgh, recognizing the strengths of the city and of CMU in helping us obtain the top talent we need," he said.
Until recently, Xing was director of the Center for Machine Learning and Health at CMU, which is part of the Pittsburgh Health Data Alliance, a collaboration between Carnegie Mellon, the University of Pittsburgh and UPMC.
Though the launch of Petuum led him to step down from that post, one of the first products Petuum plans to reveal is a method for assessing disease risk and predicting readmission rates for patients by analyzing electronic medical records and searching for patients with similar conditions. The product could provide new solutions for precision medicine and decision-making problems in the health care industry.
By Michael Cunningham
Ramayya Krishnan, dean of Carnegie Mellon University's H. John Heinz III College, shared his expertise on smart cities and mobility as a guest panelist at the Consumer Technology Association's 2017 International Consumer Electronics Show (CES) in Las Vegas.
As a member of the "Smart Cities, Smart States, Smart Mobility" panel, Krishnan spoke about the intersection of smart transportation technology, policy and effective business models that form the foundation for solutions to mobility problems that cities and communities face.
"Pittsburgh has had the visionary leaders in both the public sector and the philanthropic community who understand this three-pronged strategic approach, which has enabled the community to test and ultimately utilize many smart city applications," Krishnan said.
Krishnan cited CMU's partnership with the City of Pittsburgh and the support of the Hillman Foundation, the Heinz Endowments and the Richard King Mellon Foundation in developing and implementing smart traffic signals to shorten travel times and reduce harmful vehicle emissions as an example of technology and policy working together to create a tangible smart city solution.
"That was a big win for both the city and the university," Krishnan said.
William Eggers, executive director of the Center for Government Insights at Deloitte, LLP, moderated the panel discussion, which also included Vinn White, Acting Assistant Secretary for Transportation, John Skowron, global consulting public sector leader at Deloitte, and Anand Shah, vice president of the Albright Stonebridge Group's India and South Asia practice.
Under Krishnan's leadership, the Heinz College, in partnership with the College of Engineering, has co-led the Metro 21 initiative, which focuses on 21st-century solutions to the challenges facing metro areas. Since its launch in 2014, CMU's Metro21 initiative has funded more than 30 research projects aimed at improving the quality of life for citizens of Pittsburgh and Allegheny County.
In September 2015, Metro21 and the City of Pittsburgh were founding members of the MetroLab Network, a group of more than 35 city-university partnerships in the U.S. focused on bringing data, analytics and innovation to city government through the development of smart city applications.
By Shilo Rea
Researchers at Carnegie Mellon University are getting closer to understanding how your brain perceives faces and recognizes old friends you haven't seen in years.
In a study published in the Dec. 26, 2016, issue of the Proceedings of the National Academy of Sciences (PNAS), scientists used highly sophisticated brain imaging tools and computational methods to measure the real-time brain processes that convert the appearance of a face into the recognition of an individual. The research team is hopeful the findings might be used in the near future to locate the exact point at which the visual perception system breaks down in different disorders and injuries, ranging from developmental dyslexia to prosopagnosia, or face blindness.
"Our results provide a step toward understanding the stages of information processing that begin when an image of a face first enters a person's eye and unfolds over the next few hundred milliseconds, until the person is able to recognize the identity of the face," said Mark D. Vida, a postdoctoral research fellow in the Dietrich College of Humanities and Social Sciences' Department of Psychology and Center for the Neural Basis of Cognition (CNBC).
To determine how the brain rapidly distinguishes faces, the researchers scanned the brains of four people using magnetoencephalography (MEG). MEG allowed them to measure ongoing brain activity each millisecond while participants viewed images of 91 different individuals each having two facial expressions: happy and neutral. The participants indicated when they recognized that the same individual's face was repeated, regardless of expression.
The MEG scans allowed the researchers to map out, for each of many points in time, which parts of the brain encode appearance-based information and which encode identity-based information. The team also compared the neural data to behavioral judgments of the face images from humans, whose judgments were based mainly on identity-based information. Then, they validated the results by comparing the neural data to the information present in different parts of a computational simulation of an artificial neural network that was trained to recognize individuals from the same face images.
"Combining the detailed timing information from MEG imaging with computational models of how the visual system works has the potential to provide insight into the real-time brain processes underlying many other abilities beyond face recognition," said David C. Plaut, professor of psychology and a member of the CNBC.
In addition to Vida and Plaut, CMU's Marlene Behrmann and University of Toronto Scarborough's Adrian Nestor participated in the study.
The Natural Sciences and Engineering Research Council, the Pennsylvania Department of Health's Commonwealth Universal Research Enhancement Program and the National Science Foundation funded the research.
Using MEG and computational tools to map how the brain processes faces from sight to recognition is one example of Carnegie Mellon's strengths in combining cutting-edge cognitive neuroscience with big data and analytics. The university's BrainHub initiative, which is designed to leverage these strengths, focuses on how the structure and activity of the brain give rise to complex behaviors.
By Shilo Rea
Carnegie Mellon University has created the first and only undergraduate major in behavioral economics to meet the rising demand for trained professionals in the field from the government, nonprofit and industry sectors. The Bachelor of Arts degree in behavioral economics, policy and organizations (BEPO) begins this semester and will train students to apply psychological insights to human behavior to explain and predict economic decision-making.
Governments seek to use behavioral economics to inform public policy and improve their effectiveness. In industry, it is used to position brands, inform product design, adjust hiring and performance evaluations, motivate employees and improve the quality of employee’s decisions regarding their benefits plans.
At Carnegie Mellon, BEPO students will learn about behavioral economics at the institution responsible for pioneering the field. The late CMU Professor Herbert Simon, a Nobel Prize and Turing Award winner, coined the phrase “bounded rationality” to describe a more descriptive conception of the limits of human problem-solving ability. CMU Professor George Loewenstein co-founded behavioral economics and is a renowned expert in a wide range of subjects, including decision-making over time, bargaining and negotiations, psychology and health, law and economics, the psychology of adaptation, the role of emotion in decision-making, the psychology of curiosity, conflict of interest and "out of control" behaviors such as impulsive violent crime and drug addiction.
The BEPO major is offered through the Dietrich College of Humanities and Social Sciences’ Department of Social and Decision Sciences (SDS). It complements SDS’ highly sought after Ph.D. program in behavioral decision research, which has alumni working at Facebook, Google, Fidelity Investments, Mckinsey & Company, the newly formed Consumer Financial Protection Bureau and a number of nonprofits, startups and consulting firms.
“We have a large group of top-notch faculty in behavioral economics, and I am delighted that they will channel this expertise into this exciting, forward-looking program,” said Richard Scheines, dean of the Dietrich College. “Along with excellent undergraduate programs in economics and joint majors in economics-statistics and in economics-math, this interdisciplinary major in behavioral economics gives Carnegie Mellon a very broad and deep set of undergraduate degrees in economics.”
In addition to Loewenstein, students in the BEPO major will have the opportunity to learn from world-class behavioral economists, including Linda Babcock, a prominent expert in gender discrimination in the workplace who also focuses on behavioral labor and economics and negotiation; John Miller, whose research interests include bidding behaviors in auction markets and the dynamics of political platforms in spatial elections and of price formation in simple markets; Saurabh Bhargava, who examines barriers to enrolling in social service programs, the social and economics factors that affect happiness, factors that influence job searches and wage expectations of the unemployed; and Alex Imas, a rising star in behavioral economics who studies risk taking in financial markets, how social concerns and emotions influence decision-making and preferences and self-imposed mental accounts and risk attitudes.
“Carnegie Mellon has one of the best groups of behavioral economists in the world. We are excited to introduce this new major and involve undergraduates in our groundbreaking research,” said Babcock, head of the Social and Decision Sciences Department and the James M. Walton Professor of Economics. “Our faculty are actively partnering with numerous governments and companies to bring behavioral economics insights into these organizations, and undergraduates will participate with faculty on these applied projects.”
The BEPO curriculum is interdisciplinary and will give students a firm foundation in economics and psychology as well as how to integrate the two perspectives. Students will take calculus and have a solid grounding in quantitative methods and experimental design, both in the laboratory and field.
To celebrate the new BEPO major, the Social and Decision Sciences Department will host an afternoon program featuring behavioral economics work by CMU faculty, alumni and students and their research partners at various industry and government organizations. The event, which will be held from noon - 5 p.m., Feb. 3, in Hamburg Hall 1214, is free and open to any interested student or member of the Pittsburgh or CMU community.