News 2021

April 2021

CMU Students Among Top Computer Programmers in North American Collegiate Competition

Aaron Aupperlee

Three Carnegie Mellon University students will compete against some of the best computer programmers in North America this summer as part of the International Collegiate Programming Competition. The CMU team of Chris Lambert, Zack Lee and Andy Yang placed third in its division and third overall among the 170 teams competing in last week's North American Divisional Championships. The result qualified the team for the upcoming North American Championship in August. A good finish there could send them to the World Finals later this year. "I'm proud of how we worked together as a team," said Yang, a senior majoring in computer science. "We divided up the easier problems in a way that matched our strengths. Our speed put us ahead of many of the teams and was crucial to our third place." The three-and-a-half-hour competition took place on April 22. The CMU team competed in the Eastern Division with groups from MIT, the University of Waterloo, the University of Toronto, Harvard, Yale, Princeton, Rutgers, the University of Michigan and others. Teams from MIT and Waterloo took first and second place, respectively. During the competition, teams solved individual problems that required them to write programs of 50 to 200 lines of code. The problems challenged the competitors to think creatively and code quickly. "Often you have to come up with your own algorithm," said Daniel Sleator, a professor in the Computer Science Department and coach of the two CMU teams that competed. "It's more than just being able to knock out code. It's that you have to knock out fast, bug-free code very quickly." The third-place finishers solved nine of the 13 problems. The cranked through their first problems quickly, solving eight in less than 50 minutes. Then, after hours of work, Lee managed to crack a hard computational geometry problem, landing the team its ninth problem and propelling them into third place. "I sort of just walled myself off and worked on it until I managed to solve it and submit the correct code with 20 minutes left in the contest," said Lee, a junior majoring in math and computer science. Lambert said getting the easier part of the problem set out of the way first was their plan going into the competition. "It gave us ample time and room to approach the harder problems," said Lambert, a sophomore majoring in computer science. "Zack has a lot to be proud of in pulling through with the geometry problem given how few teams ultimately got it." Only four teams in the entire competition solved that problem. "We wouldn't have qualified to the next round without it," added Yang. Another team from CMU placed seventh in the competition. Eric Wei, Dilhan Salgado and Wassim Omais completed the same number of problems as the third-place team but took more time and more attempts to do so. The North American Championship in August could be an in-person competition, meaning the team would have to share one computer and not have access to the internet. "We will have to get used to the change in competition environment," Yang said. "I think that given our performance in this round we have a good chance of advancing to the World Finals from the next round. Hopefully we can keep improving between now and then to make it happen."

Duolingo Named to Time Magazine's List of 100 Most Influential Companies

Aaron Aupperlee

Duolingo's little green owl found itself among prestigious company this week when Time Magazine named the Carnegie Mellon University spinoff and Pittsburgh startup to its list of the 100 most influential companies for 2021. The language learning app appeared on the list alongside tech heavyweights Apple, Google, Microsoft, Amazon, Facebook and Twitter; electronic vehicle competitors Tesla, GM and VW; entertainment powerhouses Disney, Sony, Nintendo and Netflix; COVID-19 vaccine heroes Moderna, Pfizer and BioNTech, and Johnson & Johnson; and other apps like Strava, Headspace, Clubhouse and Bumble. Duolingo co-founder and CEO Luis von Ahn, a CMU alumnus and faculty member, said the company was honored by the mention. "This is a testament to our incredible team and the impact we've had in bringing high-quality, accessible education to over 500 million people worldwide. We believe that education has the power to reduce economic inequality and will continue to do our part to help build a more equitable world," von Ahn said. Von Ahn started Duolingo with one of his doctoral students, Severin Hacker, who earned his Ph.D. from CMU in 2014. Hacker is the company's chief technology officer. The pair kept the company in Pittsburgh after it spun out of CMU in 2011. The company offers language learning courses and launched its 40th language, Yiddish, this month. It also developed a low-cost test to certify the English language skills of college and job applicants. Time noted that there was a 2,000% increase in the use of the test during the pandemic, allowing more international students to apply to U.S. colleges without an in-person exam. Duolingo has grown into one of the most successful CMU spinoffs and Pittsburgh tech companies. The company was the first Pittsburgh startup to reach a $1 billion valuation in late 2019, when a round of funding brought it to $1.5 billion. Less than a year later, the company was valued at $2.4 billion — the most valued startup in Pennsylvania. In selecting its list, Time focused on companies that are shaping the future. It solicited nominations from health care, entertainment, transportation, technology and other sectors, and evaluated nominees on factors like relevance, impact, innovation, leadership, ambition and success to compile their list.

Misha Ivkov Earns Stehlik Scholarship

Aaron Aupperlee

Misha Ivkov's hands shook during his first college computer science exam. Carnegie Mellon University was a bit overwhelming, much different from high school. But Ivkov got through that first exam — after that, everything was fine, he said — and the soon-to-graduate computer science major went from fearing college during his first weeks to relishing his classes and helping other students find the same experience. "I really enjoyed taking all these classes, so I wanted to be sure other students did too," Ivkov said. "I wanted students to succeed." Throughout his time in CMU's School of Computer Science, Ivkov made helping students a top priority, whether it was in the classroom as a teaching assistant (TA), or outside it through his involvement in clubs and organizations. In recognition of his efforts, Ivkov received the Mark Stehlik Alumni Undergraduate Impact Scholarship. Now in its seventh year, the Stehlik Scholarship celebrates undergraduate students about to graduate whose reach for excellence extends beyond the classroom. Awardees have demonstrated a desire to make a difference in SCS, the field of computer science and the world around them. Ivkov also received this year's Alan J. Perlis Undergraduate Student Teaching Award. Ivkov has served as a TA for Principles of Imperative Computation, Great Theoretical Ideas in Computer Science, and Probability and Computing. He also co-developed Great Ideas in Tech Interviews and Coding Screens, a student-taught class he described as a crash course to help students survive technical interviews. Ivkov said he enjoys working with students in introductory classes. "This is the first time the students see this material. As a TA, you have the opportunity to be the peer that gets them excited about it," Ivkov said. "Then, you have the chance to see them progress through their studies." His work with students extends to the clubs he participates in. Ivkov has been one of the heads of the mentoring program in SCS4All, matching undergraduate first years with sophomores, juniors or seniors to give new students a connection with someone who has been through it before and can help guide them along the way. Ivkov, who entered CMU as a math major but soon switched to computer science, has also taken a leadership role with Carnegie Mellon Informatics and Mathematics Competition (CMIMC). As co-director his sophomore and junior years, he helped write competition problems, coordinated logistics and ensured that the competition was a success for the 400 high school students participating. "At the end of the day, it's for the high school students," Ivkov said. "They get to be excited about math." Helping younger students comes almost naturally to Ivkov. "I have a little brother," Ivkov said. "So, I've been helping him with his work for a while." After graduation, Ivkov will head to Stanford to pursue a Ph.D. in theoretical computer science. He is excited to continue researching how to apply the great theoretical ideas behind computer science that he's helped other students understand for so long.

Carnegie Mellon, Heinz Endowments Launch Center for Shared Prosperity

Brian Thornton

Carnegie Mellon University and The Heinz Endowments today announced a sweeping initiative to leverage the university's internationally recognized strengths in applied research to address longstanding barriers to equity and foster economic empowerment in the greater Pittsburgh region. The Center for Shared Prosperity aims to create a sustainable and replicable model for community-university collaboration, with a focus on deploying solutions for socioeconomic inequities and making measurable progress toward greater economic prosperity and overall well-being of residents. The Heinz Endowments has committed $30 million over six years to fund the creation and launch of the center, as well as support for initial real-world projects identified by community partners that are grounded in the community. The grant, the largest in The Heinz Endowments' history, includes funding to develop, pilot and scale region-wide interventions to identify and address structural barriers to access and opportunity. A portion of the grant will be used to establish an endowment to support the center's work in perpetuity. "As a university- and community-wide effort, the Center for Shared Prosperity will apply a comprehensive methodology to CMU's engagement across Western Pennsylvania and will leverage our unique expertise to help residents benefit from the innovation economy," said Farnam Jahanian, CMU president. "The Heinz Endowments and CMU have worked together for decades on projects that support Pittsburghers, and this new initiative will expand our community collaborations at a particularly critical moment. With both the pandemic and the rapid pace of technological change contributing to a widening opportunity gap, the solutions proposed through the Center for Shared Prosperity will help our region address societal barriers and will also serve as a model that can be replicated in communities across the country. We are grateful to The Heinz Endowments and its board for their generous support and partnership." At the heart of the initiative is a new model of collaboration that unites the expertise of both the community and the university. The recently formed Center Community Committee will be charged with identifying specific equity, economic and social justice challenges facing the Pittsburgh region that will be the focus of the center's work, including in areas such as housing, education, transportation, healthcare, technology fluency and access to capital. It will include representation from Western Pennsylvania community organizations and residents; CMU faculty, staff and students; and Heinz Endowments staff. In turn, solution-oriented working groups composed of community members and CMU participants across multiple disciplines will partner to study these issues, identify structural barriers to access and opportunity, and develop and implement social and technical innovations that address them. These working groups will harness community members' lived experiences and skills, alongside CMU experts in areas where the university leads, including data science, public policy, technology, humanities and the social sciences. The Center for Shared Prosperity stems partly from discussions between Carnegie Mellon and The Heinz Endowments that began three years ago. Convinced that the Pittsburgh region's path to a sustainable economic future depends heavily on the people, discoveries and enterprises connected with its major research universities, The Heinz Endowments was considering major support for CMU — but only if the work would intentionally include the local community and its social, environmental and economic challenges. "Around the world, a relative handful of major research institutions, Carnegie Mellon among them, are literally inventing the future, with significant global benefits and impacts," said Grant Oliphant, president of The Heinz Endowments. "But too rarely are local communities and complex social needs the real beneficiary or even the focus of the knowledge, creativity and wealth-creation flowing from these extraordinary engines of innovation. We wanted to see if Pittsburgh could reinvent that paradigm, and Carnegie Mellon — with its long history of tackling real-world problems — has risen to the challenge. "This center will put the innovation talents of one of the world's best universities in service to community, and at the same time harness the insights of community on behalf of better innovation," Oliphant added. "We also hope it will send a signal to peer institutions nationally and globally that they too have a role to play in making sure the prosperity they help create is shared more broadly and equitably. Worsening inequality and social inequity are not inevitable byproducts of innovation, only failures of intention and imagination." The initiative is ambitious in scope, speed and transparency. Multiple working groups tackling critical issues will be launched within its first year, with interventions developed and piloted in the community within months of the groups' founding. Publicly available stories that illustrate the projects' progress — using multimedia, CMU's EarthTime data visualization tool and voices from the community — will be regularly published on the web. The center plans to rely on and engage expertise from across all of CMU's schools and colleges, as well as interested alumni. "Through its unique model of collaboration in which community members and university faculty and staff work together as peers, and the development and distribution of social and technological innovations with real-world applications, the center will help to dismantle barriers to shared prosperity and equity," said Illah Nourbakhsh, the K&L Gates Professor of Ethics and Computational Technologies in the Robotics Institute, who will serve as the center's inaugural executive director. "The Center for Shared Prosperity will be laser-focused on creating direct, sustainable impact on Pittsburgh by bringing research into practice." For example, Nourbakhsh's Community Robotics, Education and Technology Empowerment Lab (CREATE Lab) and its community partners recently used EarthTime to demonstrate how high rates of mortgage application denials and sharply increasing rental prices impact the ability of vulnerable populations to live and prosper in Pittsburgh. In partnership with The Heinz Endowments and the Pennsylvania Housing Finance Agency, the team is now working on innovations to make mortgage programs more equitable, such as ensuring education debt does not disqualify individuals from accessing home financing. The Center for Shared Prosperity represents a significant step forward on the university's commitment to deeper engagement with, and economic empowerment of, the broader Pittsburgh community. It also builds on a foundation of community engagement developed over the years by countless faculty, staff, students and alumni working with, and for, CMU's neighbors. "What most excites me about the Center for Shared Prosperity is the tangible commitment from one of Pittsburgh's anchor institutions to be part of the solutions," said Jamil Bey, president and CEO of the UrbanKind Institute. The organization has worked with Nourbakhsh and is leading the collaboration behind the Equitable and Just Greater Pittsburgh platform, which is helping to guide the center's work. "By investing in community-driven goals and priorities, and providing funding to test and scale these solutions, it can create momentum that could begin to create needed structural change." "Through the Center for Shared Prosperity, I hope we can come together and show other communities that we did it," says Terri Shields, executive director of JADA House International and member of the Center Community Committee. "We may not always agree, but we'll respect each other's opinions, get done what we need to do for Pittsburgh, and set an example for other cities." While it will focus on issues of local importance, the Center for Shared Prosperity aims to become a national model for university-community engagement and the translation of research into practical actions for impact. "The issues facing Pittsburgh are perhaps unique, yet our work to find solutions can provide an opportunity to share our approach and outcomes with regions across the country facing their own challenges," Nourbakhsh said. "We hope that the center's approach and continuing evolution will help catalyze other university-community collaborations as they work to advance a more equitable future."  

SCS Remembers Adobe Founder and Ph.D. Alum Charles Geschke

Aaron Aupperlee

A creator of software that revolutionized the way people collaborate, Charles M. Geschke left his mark on Carnegie Mellon University long after earning his degree. "Chuck" Geschke, who received a Ph.D. in computer science from CMU in 1973 and continued to give back to the university throughout his life, died Friday, April 16, at 81. "An influential leader in the software industry for nearly 50 years, Chuck Geschke helped to expand the usability and accessibility of computers for a broad audience," said CMU President Farnam Jahanian. "He also embodied the best qualities of the CMU alumni and computer science communities: a commitment to his colleagues, their work, and a culture of collaboration and generosity. He will be greatly missed in the computing community and the world." Geschke co-founded Adobe in 1982 with John Warnock, a colleague from Xerox. The pair's first product was Adobe PostScript. Later innovations included Portable Document Format (PDF) technology and the powerful photo and video editing tools in Photoshop and Premiere. CMU awarded Geschke an honorary doctor of science and technology in 2013. In 2012, Chuck and Nan Geschke endowed the directorship of the Human-Computer Interaction Institute (HCII). Geschke credited his CMU education and the network of people he established at the university as an important ingredient to his success. "I admire what Carnegie Mellon does and am pleased to support the Human-Computer Interaction Institute," Geschke said in 2012. "The way people relate to technology has a great impact on their lives, so it seems to me that research in human-computer interaction is a natural place where CMU can innovate." Jodi Forlizzi, the Charles M. Geschke Director of HCII, said the institute's association with Geschke is a great honor and inspiration. "We try every day to live up to the high standards that Chuck set," Forlizzi said. Those who worked closely with Geschke, including his long-time business partner Warnock, said the pioneer will be missed. "I could never have imagined having a better, more likable or more capable business partner. Not having Chuck in our lives will leave a huge hole and those who knew him will all agree," Warnock said in a statement from Adobe. Adobe CEO Shantanu Narayen called Geschke a guide and a hero for the company and the technology industry. In an email to employees, Narayen wrote that Geschke contributed to groundbreaking software that "revolutionized how people create and communicate." "As much as his inventions changed the world, it is his focus on people, purpose and culture that has profoundly impacted each of us at Adobe," Narayen wrote. "He believed that good ideas come from everywhere in the company and that it's not only what we do but how we do it that matters most. He dedicated much of his time and talent to various philanthropies and community organizations throughout his lifetime." Nan Geschke told the Mercury News that family came first for her husband. "He was a famous businessman, the founder of a major company in the U.S. and the world, and of course he was very, very proud of that, and it was huge achievement in his life, but it wasn't his focus — really, his family was," she said. "He always called himself the luckiest man in the world." Geschke's network of friends at CMU and in the world of computer science remained strong throughout his life. "Throughout his career, Chuck remained true to his principles, serving stockholders, customers, employees and community," said Jim Morris, professor emeritus and former dean of the School of Computer Science. "I am grateful — personally and as a member of the CMU community — for Chuck and Nan's friendship and generosity." Geschke retired as president of Adobe in 2000. He served as chair of the board until 2017 and then as a board member until 2020, when he became an emeritus board member. Geschke received the Marconi Prize in 2010, the equivalent of the Nobel Prize for the field of information technology. In 2008, he was elected to the American Academy of Arts and Sciences, and in 2009 President Barack Obama awarded him the National Medal of Technology and Innovation.

Algorithm Uses Online Ads To Identify Human Traffickers

Aaron Aupperlee

Ads peddling the victims of human trafficking hide among millions of escort listings online. While identifying similar ads could be the key to taking down a human trafficking organization, the sheer volume of listings — with new ones added each day — makes the task a daunting one for law enforcement. Researchers at Carnegie Mellon University and McGill University hope to simplify that task by adapting an algorithm first used to spot anomalies in data, like typos in patient information at hospitals or errant figures in accounting, to identify similarities across escort ads. The algorithm scans and clusters similarities in text and could help law enforcement direct their investigations and better identify human traffickers and their victims, said Christos Faloutsos, the Fredkin Professor in Artificial Intelligence in CMU's School of Computer Science, who led the team. "Our algorithm can put the millions of advertisements together and highlight the common parts," Faloutsos said. "If they have a lot of things in common, it's not guaranteed, but it's highly likely that it is something suspicious." The team calls the algorithm InfoShield and will present a paper on their findings at this week's IEEE International Conference on Data Engineering (ICDE). According to the International Labor Organization, an estimated 24.9 million people are trapped in forced labor. Of those, 55% are women and girls trafficked in the commercial sex industry, where most ads are posted online. The same person may write ads for four to six victims, leading to similar phrasing and duplication among listings. "Human trafficking is a dangerous societal problem which is difficult to tackle," lead authors Catalina Vajiac and Meng-Chieh Lee wrote. "By looking for small clusters of ads that contain similar phrasing rather than analyzing standalone ads, we're finding the groups of ads that are most likely to be organized activity, which is a strong signal of (human trafficking)." Vajiac is a Ph.D. student in the Computer Science Department. Lee worked on InfoShield while a visiting student at CMU and will continue to do so when he returns to pursue his Ph.D. Other authors included fellow Tartans Faloutsos and Namyong Park, a Ph.D. student in the Computer Science Department. Reihaneh Rabbany, a former post-doctoral researcher at CMU who is now an assistant professor in the School of Computer Science at McGill, and her students Aayushi Kulshrestha and Sacha Levy collaborated on the research. The team was assisted by experts from Marinus Analytics, a spinoff of CMU's Robotics Institute that uses artificial intelligence, machine learning, predictive modeling and geospatial analysis to combat sex trafficking. To test InfoShield, the team ran it on a set of escort listings in which experts had already identified trafficking ads. The team found that InfoShield outperformed other algorithms at identifying the trafficking ads, flagging them with 85% precision. Perhaps more importantly, it did not incorrectly flag any escort listings as human trafficking ads when they were not. False positives can quickly erode trust in an algorithm, Faloutsos said. Proving their success was tricky. The test data set contained actual ads placed by human traffickers. The information in these ads is sensitive and kept private to protect the victims of human trafficking, so the team could not publish examples of the similarities identified or the data set itself. This meant that other researchers could not verify their work. "We were basically saying, 'Trust us, our algorithm works,'" Vajiac said. To remedy this, the team looked for public data sets they could use to test InfoShield that mimicked what the algorithm looked for in human trafficking data: text and the similarities in it. They turned to Twitter, where they found a trove of text and similarities in that text created by bots. Bots will often tweet the same information in similar ways. Like a human trafficking ad, the format of a bot tweet might be the same with some pieces of information changed. Rabbany said that in both cases — Twitter bots and human trafficking ads — the goal is to find organized activity. Among tweets, InfoShield outperformed other state-of-the-art algorithms at detecting bots. Vajiac said this finding was a surprise, given that other algorithms take into account Twitter-specific metrics such as the number of followers, retweets and likes, and InfoShield did not. The algorithm instead relied solely on the text of the tweets to determine bot or not. "That speaks a lot to how important text is in finding these types of organizations," Vajiac said. Despite working on algorithms for forecasting and anomaly detection for 30 years, this was the first time Faloutsos applied one to stopping human trafficking. He and the team hope their work plays a role in helping law enforcement rescue victims and in reducing human suffering. Their work continues. The team talks to experts weekly to learn more about human trafficking and efforts to end it. The more they learn, the more passion they put toward stopping it. "You see how relevant and impactful your work could be," Rabbany said. "And you see how much work there is to be done, how much room for improvement there is and how much you could bring to the table.

Mani Revists AI Grand Challenges in New Article

Aaron Aupperlee

Chess turned out to be an easy one. Translating speech in near real-time is mostly done. The accident-avoiding car? Maybe halfway there. In 1988, Raj Reddy, the Moza Bint Nasser University Professor of Computer Science and Robotics at Carnegie Mellon University, proposed an ambitious set of grand challenges for artificial intelligence to tackle in the coming 30 years. With that deadline past, Ganesh Mani, Reddy's colleague and an adjunct faculty member in the Institute for Software Research, took a look at the status of the challenges and proposed a new set of tasks in a recent AI Magazine article. "Grand challenges are important, as they act as compasses for researchers and practitioners alike — especially young professionals — who are pondering worthwhile problems to work on, testing the boundaries of what is possible," Mani wrote in "Artificial Intelligence's Grand Challenges: Past, Present and Future." In the article, Mani laid out his challenges for AI in health, wealth and wisdom, proposing a shorter timeline than 30 years to match the upped speed of innovation. The new challenges are specific enough for researchers to conceptualize solutions and broad enough to enable use in a variety of other tasks. Throughout his article, Mani scatters essays from top technologists on future applications of AI and other new challenges. Among them are Frank Chen, partner at Andreessen Horowitz; Steve Cross, retired Georgia Institute of Technology faculty member and former director and CEO in the Software Engineering Institute; Vanathi Gopalakrishnan, associate professor and director of the Intelligent Systems Program at the University of Pittsburgh; Ken Stanley, research manager at OpenAI; and Thomas Kalil, chief innovation officer at Schmidt Futures. Francesca Rossi, president of the Association for the Advancement of Artificial Intelligence and an IBM fellow, proposed in her essay an ethical challenge for AI that, in part, would require an AI system to evaluate not only its own behavior but the behavior of humans working with it. "Grand challenges can be very inspirational for researchers and practitioners. Often the path to the result is more important than the result itself. Even before the challenge is achieved, many new technical, methodologies, and general lessons can be derived; and these can be reused or adapted in other contexts, leading to advancements toward other challenges as well," Rossi wrote. Mani said the idea for the article started a couple of years ago during a chat in Reddy's office about the status of the grand challenges he laid out in 1988. Developing an AI to beat a world chess champion turned out to be relatively easy, the two of them agreed. And while Mani felt that translation technology from Google, Facebook and Microsoft had accomplished the challenge of developing a translating telephone allowing people speaking different languages to talk seamlessly, Reddy demurred, pointing out that the technology is not widely accessible, for example, to a rural farmer in an emerging economy without an internet-connected smartphone. "Raj started this thread 30 years ago," Mani said. "This article was a way of continuing that conversation and advancing the technology to work with humans for the betterment of humans."

CMU's Snakebot Goes for a Swim

Aaron Aupperlee

Carnegie Mellon University's acclaimed snake-like robot can now slither its way underwater, allowing the modular robotics platform to inspect ships, submarines and infrastructure for damage. A team from the Biorobotics Lab in the School of Computer Science's Robotics Institute tested the Hardened Underwater Modular Robot Snake (HUMRS) last month in the university's pool, diving the robot through underwater hoops, showing off its precise and smooth swimming, and demonstrating its ease of control. "We can go places that other robots cannot," said Howie Choset, the Kavčić-Moura Professor of Computer Science. "It can snake around and squeeze into hard-to-reach underwater spaces." The project is led by Choset and Matt Travers, co-directors of the Biorobotics Lab. The submersible robot snake was developed through a grant from the Advanced Robotics for Manufacturing (ARM) Institute. The project aims to assist the Department of Defense with inspecting ships, submarines and other underwater infrastructure for damage or as part of routine maintenance, said Matt Fischer, the program manager at the ARM Institute working on the project. The military has limited options for inspecting areas like a ship's hull. To do so, the Navy must either send a team of divers to the ship's location, wait until it returns to port to deploy the divers, or pull it into a dry dock — all options that take time and money. A submersible robot snake could allow the Navy to inspect the ship at sea, immediately alerting the crew to critical damage or sending information about issues that need attention back to port for use when the ship docks. "If they can get that information before the ship comes into a home port or a dry dock, that saves weeks or months of time in a maintenance schedule," said Fischer, who served in the Navy for three years. "And in turn, that saves money." Fischer, who crawled into the ballast tanks of a submarine during his service, said many sailors would gladly pass that difficult and tight duty to a robot. Steve McKee, a co-lead of the Joint Robotics Organization for Building Organic Technologies (JROBOT), a Department of Defense task force interested in technology like the submersible robot snake, said the project is a great example of a partnership between CMU, the ARM Institute, and the Department of Defense that will improve the readiness of equipment in the armed services. "The advancements being made hold great promise for helping not only the Department of Defense, but also various industries around the world," McKee said. Outside the military, the robots could inspect underwater pipes for damage or blockages, assess offshore oil rigs, or check the integrity of a tank while it is filled with liquid. The robot could be used to inspect and maintain any fluid-filled systems, said Nate Shoemaker-Trejo, a mechanical and mechatronics engineer in the Biorobotics Lab working on the submersible snakebot. "The distinguishing feature is the robot's form factor and flexibility. The smallest versions of regular submersibles are usually blocky, one-piece arrangements. The robot snake is narrow and jointed," Shoemaker-Trejo said. "The end result is that an underwater robot snake can squeeze around corners and into small spaces where regular submersibles can't go." Versions of the robot snakes have already proven useful in difficult situations. Travers led a team to Mexico City in 2017 to use robot snakes in a search-and-rescue mission after an earthquake. And a robot snake made a lasting impression on Jimmy Fallon when it climbed up his leg as a guest on NBC's "The Tonight Show With Jimmy Fallon." The robot's modular design allows it to adapt to different tasks, whether squeezing through tight spaces under rubble, climbing up a tree or slithering around a corner underwater. For the underwater robot snake, the team used existing watertight modules that allow the robot to operate in bad conditions. They then added new modules containing the turbines and thrusters needed to maneuver the robot underwater. Development progressed rapidly. The team started working on the underwater robot snake in July 2020 and by March 2021, had it swimming in the CMU pool. "I'm surprised that we made this robot work as fast as we did," Choset said. "The secret is the modularity and the people working on this technology at CMU." Additional photos and video are available through Google Drive.

Eskenazi Named International Speech Communication Association Fellow

Aaron Aupperlee

Speech processing research is at a high right now, with virtual assistants like Alexa, Siri, Google and others always listening and willing to help. But without a keen eye — or ear — for who this technology aims to assist, interest could wane, said Maxine Eskenazi, a Carnegie Mellon University researcher in the School of Computer Science who has worked on speech processing and spoken dialogue systems for decades. "We need to stop focusing on the agent and start focusing on the user," Eskenazi said. "It's only a dialogue if there are two individuals participating. If we make systems that are just fun for us to make but do not serve the user and do not help the user, then they'll stop using Alexa or Google." For her contributions to the field and decades of service to the community, Eskenazi, a principal systems scientist in the Language Technologies Institute, has been selected as a fellow of the International Speech Communication Association (ISCA). ISCA is a nonprofit organization promoting research related to speech communication science and technology. "I know they don't honor many people," Eskenazi said. "When I see the other names on the list, I am so honored." Eskenazi's work has focused on obtaining quality data through crowdsourcing and developing speech systems designed to help the user. More than a decade ago, she worked on "Let's Go," a spoken dialogue system for the Port Authority of Allegheny County. The system gave users information about what buses to take and when they should board them so they could arrive at their destination at a desired time. A new similar system aimed at seniors, GetGoiing, speaks slowly, with pauses, so people can write down the information. Eskenazi is building on that system to help people navigate large buildings, like hospitals. One project helps people find their way through UPMC Presbyterian Hospital. "Google can get you there on the bus, but it won't tell you which door to enter when you arrive," Eskenazi said. "We're trying to make things easier for people to find." Eskenazi has been involved with ISCA since it was formed in the late 1980s. She was in charge of its first newsletter and signed up the organization for Facebook, its first social media presence. Proceeds from a book she co-authored, "Advanced Social Interaction With Agents," supports ISCA's student travel fund. She will be recognized at INTERSPEECH 2021, scheduled for this summer in Brno, Czech Republic.  

SCS Faculty Awarded Five Grants Through Google's Inaugural Research Scholar Program

Aaron Aupperlee

Six Carnegie Mellon University faculty members, including five affiliated with the School of Computer Science, received grants through Google's inaugural Research Scholar Program. The program provides up to $60,000 to support the research efforts of early career professors. SCS-affiliated faculty who received grants include Pravesh Kothari, an assistant professor in the Computer Science Department; Nihar Shah, an assistant professor in the Machine Learning and Computer Science departments and in the Electrical and Computer Engineering Department; Virginia Smith, an assistant professor in the Machine Learning Department and Electrical and Computer Engineering Department; Nathan Beckmann, an assistant professor in the Computer Science Department and Electrical and Computer Engineering Department; and Sivaraman Balakrishnan, an assistant professor of statistics in the Dietrich College of Humanities and Social Sciences and an affiliated faculty member in the Machine Learning Department. Balakrishnan received a grant jointly with Aravindan Vijayaraghavan, an assistant professor in the Computer Science Department at Northwestern University. Dina El Zanfaly, an assistant professor in CMU's School of Design, also earned a grant. In 2020, Google announced the creation of the Research Scholar Program and the Award for Inclusion Research Program to bolster its support for the research community. This year, the company awarded 77 grants to 86 professors from 50 universities in more than 15 countries. Nearly half of the grants went to professors identifying as a historically marginalized group within technology.

SCS Juniors Named Goldwater Recipients

Aaron Aupperlee

Two School of Computer Science students received the prestigious Barry Goldwater Scholarship for 2021. Arvind Mahankali and Jinhyung Park were among four Carnegie Mellon University students granted the scholarship, which is awarded to sophomores and juniors who show promise as leaders in the natural sciences, engineering and mathematics. Park, a junior, studies artificial intelligence and recently worked on computer vision projects related to autonomous vehicles. He has sought to create better models of the real world by combining two-dimensional and three-dimensional inputs, which will help self-driving cars see better or improve how robots interact with their surroundings. "I think I've still only scratched the surface of what computer vision has to offer," Park said. "If AI systems have a better understanding of 3D inputs, they can have a better perception of the real world." Park decided to pursue research during his sophomore year, when he would read a paper and think about how he could improve the idea or fuse it with another field. "There was so much out there, and there was so much that people were working on, and I just wanted to find out more," Park said. "I wanted to read and not just observe this progression but be part of it." Mahankali, a junior studying computer science and mathematics, has developed efficient algorithms for machine learning problems. He said he became interested in machine learning during the Freshman Immigration Course, when he learned how adversarial examples can disrupt the intended function of an algorithm. Mahankali wanted to understand how, for example, a neural network designed to identify the animal in a given picture could be fooled into thinking a panda was a gibbon. "I thought understanding those types of questions seemed interesting and useful for improving the performance and safety of these algorithms," Mahankali said. "Safety is critical in certain applications. If self-driving cars can be fooled, we might be in trouble." While taking the Algorithms for Big Data course, Mahankali was impressed by how beautiful math techniques could be used to solve real-world problems in machine learning, inspiring him to dive deep into the subject. Awarded by the federally endowed Barry Goldwater Scholarship and Excellence in Education Foundation, the scholarship provides up to $7,500 per year for tuition, fees, books, and room and board. It is the most prestigious STEM scholarship for undergraduates. Nationally, the foundation awarded 410 scholarships for 2021. Esther Bedoyan, a junior studying electrical and computer engineering and biomedical engineering, and Ethan Lu, a junior studying mathematical sciences, also received the scholarship.

Filter News

News by Year