CMU's Zoe Rover Shows Robots Can Find Subterranean Organisms

Mission in Chile's Atacama Desert Will Aid Search for Life on Mars

An autonomous rover named Zoë, designed and built by Carnegie Mellon University's Robotics Institute, drilled into the soil of Chile's Atacama Desert in 2013 and discovered unusual, highly specialized microbes. The NASA-funded mission demonstrated how robots might someday find life on Mars. The astrobiology mission was led by the Robotics Institute and the SETI Institute to test technologies for searching for life underground. The microbial analyses of the soil samples recovered by Zoë were published Feb. 28 in the journal Frontiers of Microbiology. "This experiment culminates more than a decade of research at Carnegie Mellon developing robots that can autonomously explore the geology and microbiology of planetary surfaces," said David Wettergreen, research professor of robotics and principal investigator of the Life in the Atacama project. "Beginning in 2003, Zoë mapped more than 300 kilometers of traverse at eight field sites in the Atacama." "We have shown that a robotic rover can recover subsurface soil in the most Mars-like desert on Earth," said Stephen Pointing, professor of environmental studies at Yale-NUS College, Singapore, who led the microbial research. "This is important because most scientists agree that any life on Mars would have to occur below the surface to escape the harsh surface conditions where high radiation, low temperature and lack of water make life unlikely."

Hammer, Ogan Accept Moran Professorships

Jessica Hammer and Amy Ogan, faculty members in the Human-Computer Interaction Institute (HCII), received the inaugural Thomas and Lydia Moran Career Development Professorships in Learning Science during a Feb. 19 ceremony. The Morans, who both have ties to Carnegie Mellon, recently endowed the professorships to support junior faculty members dedicated to the study and improvement of learning. Hammer, an assistant professor who has a joint appointment with the Entertainment Technology Center, studies the psychology of games and how specific game design decisions affect how players think and feel. She designs games intended to change lives for the better. Her Oh!Lab works at the intersection of culture, learning, play and design to create new interactions and experiences. She earned her bachelor’s degree in computer science at Harvard University, a master’s degree in interactive telecommunications at New York University and a Ph.D. in cognitive studies in education at Columbia University. Ogan studies how to make learning experiences more engaging, efficacious and enjoyable. Her research contributes to learning theory and informs the design of next-generation educational technologies, supporting both social and cognitive aspects of learning. She earned both her bachelor’s degree in computer science and her Ph.D. in human-computer interaction at Carnegie Mellon. She was a post-doctoral fellow in HCII before joining the faculty as an assistant professor in 2014. Thomas Moran earned his Ph.D. in computer science at Carnegie Mellon in 1974 and is the first graduate student to receive a degree in human-computer interaction. A pioneer in HCI and founder of the journal Human-Computer Interaction, he managed user-interface and collaborative-systems research at Xerox, and later was a distinguished engineer at IBM Almaden Research Center. Lydia De Benedetti Moran is a graduate of Oberlin College and a long-time volunteer and supporter of educational charities. Her late father, Sergio De Benedetti, was a CMU physics professor, and her brother, sister and an uncle are CMU alumni.

CMU-Developed Software Aids Sting Operation Against Sex Traffickers

Ad-tracking software developed by the Robotics Institute's Auton Lab and spinoff company Marinus Analytics helped the FBI and local law enforcement agencies in more than a dozen U.S. cities execute a sting operation targeting illegal Asian brothels. The software, called Traffic Jam, was used by the Pittsburgh-based National Cyber Forensics and Training Alliance (NCFTA) to support the investigation, which resulted in a federal grand jury in Oregon charging six people with sex trafficking. The indictments were unsealed last month following a series of arrests by FBI Portland's Child Exploitation Task Force during the federal government shutdown. "Pittsburgh has been at the forefront of developing modern approaches to combating the hidden crime of sex trafficking," said Cara Jones, CEO and co-founder of Marinus Analytics. "Private-public partnerships like these have produced effective methods and tools to keep law enforcement ahead of the curve in the digital era." According to the indictments, a criminal enterprise headed by Zongtao "Mark" Chen recruited women, primarily in China, to travel to the United States to engage in prostitution. The women worked at brothels set up in hotels and apartment complexes. The Traffic Jam software aids law enforcement in identifying sex trafficking activity by analyzing images and text in online ads for escort and other sex-related services. It was originally developed by Artur Dubrawski, the research professor who heads the Auton Lab; Emily Kennedy, a Dietrich College of Humanities and Social Sciences alumna and a former research analyst in the Auton Lab; and other researchers. Kennedy, co-founder and president of Marinus, is widely recognized for her work with Traffic Jam, including being named to Forbes magazine's 2019 list of 30 Under 30 Social Entrepreneurs and earning a 2018 Toyota Mother of Invention award. The NCFTA, which identified the criminal group's cyber presence and assisted in the seizure of websites related to the Chen investigation, is a nonprofit partnership between private industry, government and academia to enable collaboration in identifying, mitigating and disrupting cybercrime.

New Method Identifies Which Asthma Patients Respond to Systemic Corticosteroids

Computational Tool Could Help Relieve Suffering for Patients Not Helped by Mainline Therapy

Physicians will be able to predict which of their patients with severe asthma are likely to benefit from treatment with systemic corticosteroids — and which might only suffer their side effects — with help from a dozen clinical variables researchers have identified using machine learning techniques. Physicians already have some clues about which patients are most helped by corticosteroid injections or pills. But the newly identified set of variables — when processed by computer software — will yield more precise predictions of a patient's response, said Wei Wu, a faculty member in Carnegie Mellon University's Computational Biology Department. "Systemic corticosteroids are the most effective therapy we have for asthma, but not all patients respond in the same way," Wu said. "Unfortunately, when clinicians don't see a big improvement after initial treatment, they might give patients even higher doses. If a patient is one of those who can't be helped by corticosteroids, the higher dose just means worse side effects." The study, led by Wu and Dr. Sally E. Wenzel, director of the University of Pittsburgh Asthma Institute at UPMC, was recently published online by the American Journal of Respiratory and Critical Care Medicine. Asthma affects about one in every 12 Americans and the rate continues to rise. The lifelong disease causes wheezing, breathlessness, chest tightness and coughing. Predicting how people will respond to corticosteroid therapy could significantly reduce the suffering of many patients, Wenzel said. "I see so many patients in my clinic who have been ravaged by the side effects of corticosteroids," said Wenzel, also chair of the Department of Environmental and Occupational Health at Pitt's Graduate School of Public Health. Weight gain, extreme emotions, inability to sleep, glaucoma and thinning of the skin are among the possible side effects of corticosteroid pills and injections, so physicians would like to prescribe them only to patients they know will benefit from them, she added. She emphasized that the study addresses corticosteroid pills and injections, not the widely used corticosteroid inhalers, although there is likely to be some overlap in patient response to the medications in either form. To better understand how different subgroups of patients respond to systemic corticosteroid therapy, the researchers used a machine learning algorithm to sift through 100 variables for each of 346 adult patients in the federally funded Severe Asthma Research Program (SARP). The algorithm, developed by Wu and Seojin Bang, a Ph.D. student in CMU's Computational Biology Department, recognizes patterns in massive volumes of complex clinical data. It clustered patients into four subgroups, including two for severe asthmatics — one that responded to systemic corticosteroids and one that didn't. Of the original 100 variables, they identified 12 — including age of onset, weight, race and scores on a quality-of-life questionnaire — that could correctly categorize patients with high confidence if processed by a computer app. To test this process, they used the 12 variables (or their equivalents) to categorize a group of 182 SARP participants not included in the original analysis. The variables proved effective in successfully categorizing these additional patients. The benefits of systemic corticosteroids can be substantial, so physicians likely will continue to try them initially in the treatment of severe asthma, Wenzel said. But once software becomes available for practitioners to predict patient response, she said they will likely switch to alternative therapies, rather than increase corticosteroid dosages, if patients haven't responded and fall into the subgroup of patients that don't usually benefit from the drugs. "We believe we've made progress toward making precision medicine a reality," Wei said. "Five years ago, we were only able to categorize patients clinically. Now, using incredibly complex data, we're able to predict how these subgroups will respond to a critical drug treatment." In addition to Wei, Bang and Wenzel, authors include other investigators in the SARP network, which is sponsored by the National Heart, Lung and Blood Institute.

CyLab Study: How the Twitter Community Stopped Fake News About Oscar-Nominated "Black Panther"

On Feb. 16, 2018, mere hours after Academy Award-nominated "Black Panther" first released to theaters, fake news trolls and malicious social media bots went to work. "My friend and I went to the #BlackPanther premier and he was brutally beaten for 'not belonging there' by an angry group who did not have tickets. Very sad, night ruined," one account posted on Twitter, including an unrelated photo of an domestic abuse victim. The incident never happened, and the account was later suspended. CyLab's Kathleen M. Carley and some of her graduate students watched as this and many other similar fake stories of racially motivated attacks swirled around Twitter during Black Panther's opening weekend. "We believe this was the first time that anyone captured the full timeline of the spreading of fake news," says Carley, a professor in School of Computer Science's Institute for Software Research (ISR). In their study, "Beaten Up on Twitter? Exploring Fake News and Satirical Responses During the Black Panther Movie Event," Carley and her students explored the differences and relationships between nonsatirical "fake news" and satirical responses on Twitter. In their analysis, the researchers collected all tweets containing "#BlackPanther" posted up to a week before and a week after the movie's theatrical release. In addition to the nonsatirical fake stories, the group found a number of satirical fake news tweets – ones in which users reposted the original fake stories but included images from cartoons, movies or classical art that depicted unrealistic violence or unrelated content. "Some users were using satire, saying things like, 'I went to the movie and I got beaten up,' and including a photo of a SpongeBob SquarePants beaten up," Carley says. "We think this actually helped discredit the original fake story post." In the two-week period, the researchers found 249 fake tweets from 238 unique Twitter handles. Of those, 178 were labeled as "satire" while 71 were labeled as nonsatire. Using an algorithm called "CMU Bothunter," the researchers found that 14 of the 238 accounts (9 percent) were bots. Accounting for all retweets and replies, the dataset of all satirical and nonsatirical fake stories totaled nearly 300,000 out of an overall data set of over 5 million tweets about Black Panther in the two-week period. CMU Bothunter estimated over 150,000 of those tweets were generated by bots, with nearly 3,000 of the bot-tweets involving fake news. The team noticed that a peculiar thing would happen after a nonsatirical fake story post was followed by a satirical post. "We were able to see that fake news stories could be stopped by counter stories that were basically satire," Carley says. "A fake story would go up, it would be spread through retweets, then a satirical story would go up, and the spreading of the first story would stop." Carley says they noticed one more way in which the spreading of fake news was stopped, and she suggests that everyone should employ it. "We saw that fake news could be stopped by people calling it out," she says. "People would quote the original message and say, 'That didn't happen.'" The researchers say that the number of tweets calling-out the fake news was greater than the fake news tweets themselves. This power-in-numbers helped halt the spread of the fake news tweets. Other authors in this study included ISR post-doctoral associate Matthew Babcock and ISR Ph.D. student David Beskow.

Haeupler, Mohimani Receive Sloan Research Fellowships

Bernhard Haeupler, assistant professor in Carnegie Mellon University's Computer Science Department, and Hosein Mohimani, assistant professor in the Computational Biology Department, are among 126 recipients of 2019 Sloan Research Fellowships, which honor early career scholars whose achievements put them among the very best scientific minds working today. The new Sloan fellows also include Florian Frick, assistant professor of mathematical sciences. A full list of the 2019 fellows is available on the Sloan Foundation website. "Sloan Research Fellows are the best young scientists working today," said Adam F. Falk, president of the Alfred P. Sloan Foundation. "Sloan fellows stand out for their creativity, for their hard work, for the importance of the issues they tackle, and the energy and innovation with which they tackle them. To be a Sloan fellow is to be in the vanguard of 21st century science." Mohimani, who joined the faculty in 2017, uses high-powered computation to unearth previously undiscovered antibiotics. He has led international teams of scientists to develop methods for rapidly searching massive databases to discover novel variants of known antibiotics, identifying compounds that hold antibiotic promise and could help fight antibiotic resistance. To discover new classes of antibiotics, his team is developing new techniques for rapidly searching large mass spectral datasets for hypothetical antibiotic molecules predicted in the genome of microbes. Mohimani earned his Ph.D. in electrical and computer engineering at the University of California, San Diego, where he was a project scientist before joining CMU. Haeupler, who joined the faculty in 2014, explores how to achieve efficient, reliable communications and prevent communication bottlenecks as future computer networks become increasingly decentralized. His research combines and goes beyond the traditional boundaries of theoretical computer science, information theory, and the theory of distributed computing, and builds foundations for the technological revolutions of the future. Haeupler earned his Ph.D. in computer science at MIT, where his dissertation won the George Sprowls Award for best computer science Ph.D. thesis and the 2014 ACM-EATCS Doctoral Dissertation Award for Distributed Computing. Last year, he received a National Science Foundation CAREER Award for early career scientists. He was a post-doctoral researcher at Microsoft Research Silicon Valley before joining CMU. The Sloan Fellowships are open to scholars in eight scientific and technical fields — chemistry, computer science, economics, mathematics, computational and evolutionary molecular biology, neuroscience, ocean sciences, and physics. Winners receive a two-year, $70,000 fellowship to further their research.

Carnegie Mellon and Lockheed Martin Sign Research Agreement

Master Agreement Will Strengthen CMU's Long-Standing Ties With Company

Carnegie Mellon University and Lockheed Martin have entered into a new master research agreement that will guide future joint research projects and enable the organizations to respond quickly to new opportunities. Lockheed Martin and its Sikorsky subsidiary have sponsored research and supported student groups at Carnegie Mellon since 1986, including a wide range of projects in the School of Computer Science and College of Engineering. Like a number of leading companies, Lockheed Martin recently has expressed interest in partnering with Carnegie Mellon on research into artificial intelligence, which President Trump declared a national priority in an executive order signed last week. "Carnegie Mellon and Lockheed Martin have enjoyed a long, productive relationship, and this new master research agreement will help us expand and accelerate collaborations between our organizations on a range of important research topics," said Michael McQuade, CMU's vice president for research. "We are especially pleased that Lockheed Martin is joining our CMU AI ecosystem, advancing a technology that will be critical for our nation's welfare." In one recently approved project, Lockheed is supporting work at CMU on how AI can enhance multiagent decision-making for missions such as humanitarian assistance and disaster relief. Researchers are focusing on how to improve coordination in such settings when not all of the players have the same information or may not be fully cooperative. "Lockheed Martin and CMU have a long and successful history of working together to solve some of the world's most complex technical challenges," said Lockheed Martin Vice President for Technology Strategy and Innovation Robie I. Samanta Roy. "Lockheed Martin is making significant investments in AI, so expanding our research partnership with CMU was a natural next step that will help us to continue accelerating the pace of innovation and create next-generation and generation-after-next technologies." In preparation for the new research agreement, Lockheed Martin executives have visited CMU's campus to discuss potential AI and other research opportunities. "AI will have a pervasive impact on our lives, from improving the quality of healthcare to creating smarter cities with less congestion and pollution," said Tom Mitchell, interim dean of the School of Computer Science. "AI advances over the past decade have been significant, but the coming decade promises to be even more exciting, given the increased R&D effort by companies such as Lockheed Martin."

Kiesler Elected to National Academy of Engineering

Sara Kiesler, Hillman Chair Emerita of Computer Science and Human-Computer Interaction in the Human-Computer Interaction Institute, has been elected to the National Academy of Engineering, among the highest professional distinctions accorded to an engineer. Her citation is "for leadership, technical innovation, and identification of social trends with the adoption of computers and robots in work and society." She is among 86 new members and 18 foreign members announced today by NAE President C. D. (Dan) Mote Jr. Kiesler has served as a program director in the division of social and economic sciences at the National Science Foundation since 2016. Kiesler's research has revolutionized the field of human-computer interaction (HCI). She has focused on many of computing's most significant social impacts, including open communication, information sharing and distributed collaboration. She also has brought concepts from social psychology and HCI to robotics, helping to create the new interdisciplinary field of human-robot interaction. Also among the newly elected members is Monica S. Lam, a professor of computer science at Stanford University who earned her Ph.D. at CMU's Computer Science Department in 1987. Her NAE citation is "for contributions to the design of advanced compiler and analysis systems for high-performance computers." NAE membership honors those who have made outstanding contributions to "engineering research, practice or education, including, where appropriate, significant contributions to the engineering literature" and to "the pioneering of new and developing fields of technology, making major advancements in traditional fields of engineering, or developing/implementing innovative approaches to engineering education." Individuals in the newly elected class will be formally inducted during a ceremony at the NAE's annual meeting on Oct. 6 in Washington, D.C.

Scaling Up Search for Analogies Could Be Key to Innovation

AI, Crowdsourcing Can Identify Insights That Lead to Breakthroughs

Investment in research is at an all-time high, yet the rate of scientific breakthroughs isn't setting any records. To resolve this quandary, scientists are turning to artificial intelligence and crowdsourcing for help in identifying a key inspiration for innovation — the perfect analogy. Wilbur Wright, for instance, famously got his idea for using wing warping to steer an airplane while twisting a cardboard box. Using similar methods to solve disparate problems is a common theme in the history of innovation. But as problems become more complex and the amount of scientific information explodes, finding helpful analogies can be difficult, said Niki Kittur, a professor in Carnegie Mellon University's Human-Computer Interaction Institute. As described in a new report to be published online this week by the Proceedings of the National Academy of Sciences, researchers are addressing this problem by breaking down the process of identifying analogies, using crowd workers to solve individual steps in the process and training AIs to do part of the work automatically. "We're developing new tools that could unlock a whole set of interesting possibilities," said Kittur, the lead author. "We're just beginning to see how people might use them." If this approach proves successful, researchers need not rely on a lone genius such as Wright to find analogies. Instead, they can use a mix of individuals and AIs, each doing those portions of the work that leverage their particular strengths, said the authors, who include scientists from CMU, the Bosch Research and Technology Center in Pittsburgh, the Hebrew University of Jerusalem, the University of Maryland and New York University Stern School of Business. Coordinating those efforts can be a challenge, they acknowledge, but better analogies could yield more efficient scientific discovery, potentially making scientific advances more profound and less incremental. "People are really interested in how we start generating breakthroughs again," said Dafna Shahaf, assistant professor of computer science at Hebrew University of Jerusalem. "The pace of discovery is high, but does not scale with the amount of resources invested in research." People, such as crowd workers on Amazon Mechanical Turk, have been key to the research, though AI can learn from their efforts and assume a larger role moving forward. For instance, the authors developed an AI tool that enables a designer to specify a focus of a product description and then abstract it in a targeted manner. A designer developing an adjustable soap dish, for example, could identify the focus as an extendable product for different sizes of soap. The focus could then be broadened to include different types of personal products or to accommodate dimensions such as heights or weights, rather than just length. The researchers have shown how this approach can be extended to scientific research. That includes developing methods for novices to annotate scientific literature, which can be challenging to read and understand. Even so, nonexperts often can discern where the most important concepts and mechanisms are in these research reports, even if they don't grasp what those concepts/mechanisms mean, said Joel Chan, assistant professor of information studies at the University of Maryland. "Knowing which parts are important buys us a lot in terms of finding subtle analogical relationships between research papers," Chan added. For example, once nonexperts isolate the parts of papers that describe their purpose or research goal, AI models can identify other papers that are about common purposes, even if they are from different topic areas. If analogy identification can be scaled up, the potential for advances is great, said Hila Lifshitz-Assaf, assistant professor of information, operations and management sciences at NYU Stern. Waiting to be tapped are more than 9 million U.S. patents; more than 2 million product and solution ideas submitted to ideation platforms such as InnoCentive, Kickstarter, Quirky and OpenIDEO; hundreds of millions of scientific papers and legal cases searchable on Google Scholar; and billions of webpages and videos searchable on the internet. Of course, the sheer volume of that information poses a challenge to finding and applying analogies, one of three challenges the authors identify. Another is the tendency of people to fixate on surface-level details, rather than deeper concepts that apply across fields. People considering how to treat an inoperable tumor with radiation without destroying healthy tissue, for instance, tend to focus on radiation or cancer rather than drawing inspiration from military science for multipronged assaults. A third challenge is the sheer complexity of real-world problems, which might require solutions of several subproblems, requiring multiple analogies at multiple levels of abstraction. Solving those challenges could usher in a new era of discovery, Kittur said, providing people with the inspiration necessary to make breakthroughs now just beyond our reach. "It could be that the low-hanging fruit has been plucked and we just don't have the ladders to reach what remains," he explained. "AI will help us get higher into the tree, but you'll still need people to actually pick the fruit." The National Science Foundation, Bosch, Google, the Israel Science Foundation, the HUJI Cyber Security Research Center and the Industrial Research Institute supported this research.

Carnegie Mellon Hosts Activation of U.S. Army AI Task Force

The United States Army is activating its Artificial Intelligence (AI) Task Force at the birthplace of AI itself: Carnegie Mellon University. The activation, which will take place on Feb. 1 at Carnegie Mellon's National Robotics Engineering Center (NREC), augments the Army's long-standing commitment to modernization and future technology, while also strengthening its ties to fundamental research in academia. CMU is serving as the hub of the AI Task Force, which will eventually include other leading universities from across the country as well as engagement with the private sector. The Task Force will also leverage close coordination and extensive collaboration with the Army Research Lab and the Department of Defense's newly established Joint Artificial Intelligence Center. "The launch of this national network based at CMU represents an important effort for the United States Army and our nation," said Carnegie Mellon President Farnam Jahanian. "CMU looks forward to working closely with our partners to ensure this robust network of AI collaborators flourishes and benefits from the long-standing strengths in AI at Carnegie Mellon and the Pittsburgh region." The AI Task Force will allow the Army to better connect with the broader AI community, which includes top-tier research universities and American companies. The technology they develop together will modernize processes used to equip and protect soldiers, enhance readiness and increase the Army's capabilities. "The activation of the Army's Artificial Intelligence Task Force is a critical step in modernizing the force to ensure future success," said Secretary of the Army Mark Esper. "Carnegie Mellon and the Pittsburgh area embody the spirit of hard work and innovation essential to shaping the Army of the future." The Task Force will be overseen by the newly established Army's Futures Command (AFC). The AFC marks the most significant Army reorganization effort since 1973, when Forces Command and Training & Doctrine Command were established. The Task Force's initial priorities are for applications of artificial intelligence to equipment maintenance/logistics; situational awareness; and humanitarian assistance and disaster relief. "The Army's AI Task Force is a good example of our commitment to capitalize on the spirit of American ingenuity found on today's colleges and universities, including Carnegie Mellon," said AFC Commander General John Murray. "Army Futures Command will posture the Army for the future by setting strategic direction, integrating the Army's future force modernization enterprise, aligning resources to priorities and maintaining accountability." Representatives from nearly a dozen universities will attend the activation ceremony as invited guests of the Army. They include Case Western Reserve University, Georgia Institute of Technology, Massachusetts Institute of Technology, Penn State University, University of Illinois, University of Maryland, University of Pittsburgh, University of Southern California, Texas A&M University, University of Texas, Austin and the University of Texas System.