In the News
University of Notre Dame computer scientists within the Lucy Family Institute for Data & Society and the College of Engineering have received an award of $300,000 from the National Science Foundation (NSF) to support the development of a new machine learning framework used to offer personalized dietary recommendations to address the national opioid epidemic.
Chuxu Zang, assistant professor of Computer Science at Brandeis University, will be part of a team using the data generated from online crowd-sourcing and business review platforms such as Yelp, to develop a dietary standard which incorporates multifactorial conditions, such as diet preference and nutrient diversity. The personalized dietary guide will be directly available to recovering opioid users, and can also aid clinicians in developing recovery care plans.
Peizhao Li, a current PhD student in the computer science department, has been awarded an NIJ (National Institute of Justice) Graduate Research Fellowship, for his proposal, "Regulating Data Bias in Intelligent Decision Making." The fellowship with a grant of $55,500 will support his research on AI fairness. This is a prestigious award which has been given to scholars doing cutting edge research in the field.
Dylan Cashman, joined the department as an assistant professor with a focus on data science and advanced visual analytics for machine learning. Previously, Dylan was a senior expert in data science and advanced visual analytics in the Data and AI division at Novartis in Cambridge, MA. He also volunteered as a web chair of the organizing committee for IEEE VIS, the premier conference for visualization research. He received his PhD from Tufts University.
Subhaheep Sarkar joined the department as an assistant professor bringing with him an expertise in data structures and algorithms, data management, storage engine, systems, data privacy, and cloud computing. Before joining Brandeis, Subhadeep was a post-doctoral associate at Boston University in the Data-intensive Systems and Computing (DiSC) lab with Manos Athanassoulis (2019-2023). Prior to that, he spent two years at Inria, Rennes (France) as a post-doctoral researcher with Christine Morin (2017-2018). He received his PhD in Computer Science from IIT Kharagpur.
Professor Harry Mairson has been selected as the Rieman and Baketel Fellow for Music for the 2023-24 academic year at Harvard-Radcliffe Institute. The Radcliffe Fellowships are annually awarded to a select group of scholars in the humanities, sciences, social sciences and arts, as well as writers, journalists and other distinguished professionals. Read the full article in BrandeisNow.
Professor James Pustejovsky was just awarded a grant for $149,997 from the NSF for his proposal entitled: "EAGER: Integrating Dense Paraphrased-Enriched Representations with Large Language Models."
Brandeis University will receive a subgrant of $825,000 as part of a $16 million grant awarded by the Mellon Foundation to the WGBH Educational Foundation in support of the American Archive of Public Broadcasting. The grant continues a collaboration between James Pustejovsky's Research Lab for Linguistics and Computation and the GBH Archives.
The Brandeis Lab will further develop and improve the set of open-source tools and workflows known as CLAMS (Computational Linguistics Applications for Multimedia Services). The work will mainly focus on applying CLAMS to a substantially larger and more diverse dataset by providing workflows and methodologies that allow archivists to adapt current AI and CL tools to new data. Pustejovsky is the TJX Feldberg Professor of Computer Science at Brandeis. Read the article published in BrandeisNow.
Professor Nianwen Xue and James Pustejovsky were awarded a $999,689 NSF Grant for a project entitled "Building a Broad Infrastructure for Uniform Meaning Representations." When humans attempt to talk with a computer, our language needs to be translated into a meaning representation that can be processed and understood by the computer. Currently, such translation is done on a task-by-task and language-by-language basis.
Such a fragmented approach introduces redundancy and repetition, and is thus inefficient. Uniform Meaning Representation (UMR) is designed as a machine-readable language that all languages, from high-resource languages such as English and Chinese, to low-resource languages like Arapaho, can be translated into. UMR can also be extended to multi-modal settings to represent the content of videos and images, allowing computers to better process and understand the content of these media forms. This project aims to build the necessary infrastructure for translating languages and other media into UMRs.
A paper entitled "Multimodal Semantics for Affordances and Actions" authored by Professor James Pustejovsky and Nikhil Krishnaswamy has been selected to receive the Best Paper Award of the HCI 2022 Thematic Area. The selection process has been performed by a review committee coordinated by the Chair of the HCI 2022 Thematic Area. The announcement of the winners will be made during the Opening Plenary Session on June 26.
A research paper entitled "Deep Learning Assisted Investigation of Electric Field-Dipole Effects on Catalytic Ammonia Synthesis" was recently accepted by JACS Au, a peer-reviewed open access multi-disciplinary journal focusing on traditional core fields. The paper involved the work of a PhD candidate in Computer Science, Han Yue (co-first author), Professor Hongfu Liu from the Computer Science department, and collaborators from the University of Massachusetts Lowell.
This paper used density functional theory (DFT) calculations — assisted and accelerated by a deep learning algorithm — to investigate the extent to which ruthenium-catalyzed ammonia synthesis would benefit from the application of such external electric fields.
Computer Science Professor Harry Mairson’s focus outside the classroom is on building violins, violas and violoncellos. At the same time, he's trying to understand and share the "secrets of Stradivari" and those of other great Italian makers of renowned classical stringed instruments. He's brought his digital tools to this work together with his workshop's traditional hand tools. Read the full article in BrandeisNow.
A research paper entitled "Economic Recession Prediction Using Deep Neural Network," was recently accepted by The Journal of Financial Data Science, a peer-reviewed journal focuses on big data analytics, artificial intelligence and machine learning in the financial industry and quantitative asset management.
The paper involved the work of two Brandeis computer science students, Zihao Wang (undergraduate, first author), Kun Li (master, second student), Professor Steve Xia from the International Business School and Professor Hongfu Liu from the Computer Science department. The paper identifies a deep learning methodology of Bi-LSTM with Autoencoder as the most accurate model to forecast the beginning and end of economic recessions in the U.S and provides good out-of-sample predictions for the past two recessions and early warning about the COVID-19 recession.
A research paper entitled "RxNet: Rx-refill Graph Neural Network for Overprescribing Detection" recently won the best paper award at CIKM 2021, a top data mining conference. This paper involved the work of Professor Chuxu Zhang of the Brandeis Computer Science department and collaborators from CWRU and Notre Dame. This paper proposed a dynamic heterogeneous graph neural network for patients' overprescribing behavior detection.
Computer science professors James Pustejovsky and Nikhil Krishnaswamy of Colorado State University have spent the past few years developing the Diana prototype in their multi-multimodal simulation work, which exists today as an interactive avatar on a computer, iPad or iPhone.
Diana is a teaching assistant designed to respond to students' nonverbal and visual cues in an effort to help middle-school teachers run their classrooms more smoothly. As students interact in the classroom, the goal is for Diana to notice students' facial expressions, conversations, gazes and gestures to infer whether they could use help or are getting distracted. Diana then responds by engaging them in conversation or prompting the teacher.
The tool could be especially useful when a teacher divides students into small groups and can only focus on one at a time, Dr. Pustejovksy says. Next, the researchers want to teach Diana to reliably recognize faces and voices, particularly diverse skin colors, accents and local dialects — a common oversight in data collection that can lead to gaps in AI's effectiveness. They plan to strengthen Diana's capabilities by collecting more visual and audio recognition data from student interactions in five middle schools in Colorado this fall. Read the full article in the WSJ on Brandeis Now (subscription needed) as well as the Fox5 New York television coverage.
A research paper entitled "Deep Clustering-based Fair Outlier Detection," recently accepted by SIGKDD 2021, a top data mining & Machine Learning conference, involved the work of Brandeis undergraduate, Hanyu Song (first author), a PhD candidate in Computer Science, Peizhao Li (second author) and Professor Hongfu Liu from the Computer Science department. This paper proposes a method for a newly emerged fair outlier detection problem.
Professor James Pustejovsky was awarded a $400,000 NSF Grant for his project entitled, "Elements: Towards a Robust Cyberinfrastructure for NLP-based Search and Discoverability over Scientific Literature." This project creates an open platform for accessing and mining information from scientific texts that provides access to an array of software, computing resources, and publication data.
Current search technologies typically find many relevant documents, but do not extract and organize the information content of these documents or suggest new scientific hypotheses based on this organized content. Natural Language Processing (NLP) strategies are a recognized means to approach this problem, and this project develops the cyberinfrastructure to support sophisticated search and retrieval from scientific publications, use and augmentation of facilities for advanced and well-established natural language processing and machine learning tools, and extraction and aggregation of data from scientific publications. See information on the NSF Grant.
A paper recently published in Cell Press, Patterns, a prestigious journal in the field of data science in interdisciplinary research, was the work of two Brandeis faculty members and two Brandeis undergraduates. The paper entitled, "Network-based Virus-host Interaction Prediction with Application to SARS-CoV-2," was co-authored by undergradates Feng Chen and Hangyu Du, and&# Computer Science Professors Hongfu Liu and Pengyu Hong as corresponding authors.
The machine learning-based model proposed by the authors in the paper provides the science community with crucial and novel insights to combat SARS-CoV-2. Read more about the publication.
Researchers at the Materials Research Science and Engineering Center (MRSEC) are harnessing the power of swirling cellular proteins to create self-propelling fluids and have taken significant steps toward understanding the swirl and controlling its flow. In a November paper in the journal Soft Matter ("Machine Learning Forecasting of Active Nematics"), professor of computer science Pengyu Hong, Fraden and Hagan demonstrated how computers could learn purely from experimental data to predict the emergence and behavior of swirl defects using deep learning techniques.
One day, self-propelling liquids could be used to create a class of liquid drugs that are injected into the bloodstream and then autonomously flow toward a specific group of cells or organs. Because the swirl is made from components already in our cells, it wouldn’t be rejected by our bodies.This is another important step towards understanding the swirl’s overall motion. Read the full article here in BrandeisNow.
On Nov. 20, 2020, the department hosted a special guest speaker, Adam Cheyer, '88, who presented "A Brandeisian Entrepreneur's Journey." Adam is the co-founder of Siri and Bixby, and founder of change.org. He spoke about harnessing what he learned at Brandeis to make a profound and positive impact on the world.
The Department welcomed two new computer science professors this fall. Iraklis Tsekourakis, associate professor completed his undergraduate degree at Aristotle University of Thessaloniki in Greece. He joined the Department of Computer Science at Stevens Institute of Technology as a PhD student in 2012 and upon completing his doctorate took up appointment as an assistant teaching professor there. His interest in teaching is complemented by research interests in computer vision and dynamic 3D reconstruction, the topic of his PhD dissertation.
Associate Professor Chuxu Zhang's research interests include data science, applied machine learning and artificial intelligence. Upon receiving his master’s from Rutgers University in 2017, he joined the Department of Computer Science and Engineering at the University of Notre Dame where he recently received his PhD. Recent projects include developing machine learning tools to solve recommendation problems in heterogeneous networks, and applying artificial intelligence to natural language processing and to synthetic chemistry.
With support from the Robust Intelligence program in the Division of Intelligent and Information Systems (IIS) and the NSF 2026 Fund Program in the Office of Integrated Activities, investigators at Boston College and Professor James Pustejovsky's team at Brandeis University are addressing the challenge of creating Artificial General Intelligence by synthesizing symbolic or logical reasoning, learning through interaction with the environment, as well as state-of-the-art neural networks. Inspired by the structure of natural (e.g., human) intelligence, the resulting mental architecture deploys each of these strategies for the problems they excel at (the "Best of All Worlds?, or BAW, approach).
Successful completion of this project will facilitate a range of research projects in AI and psychology/neuroscience. Long-term, the development of AGI is expected to have significant benefits to society, by enabling computers to develop abstract concepts grounded in experience with the world, and to generate novel ideas and inventions. This project will also help broaden student training and participation of women and underrepresented minorities. Read Professor Pustejvosky's interview with BrandeisNow!
This project aims to prototype a new architecture and test it against an open-ended task that is difficult for artificial intelligence but mastered by human toddlers everywhere: uncovering the affordances of blocks, containers, and other small objects. The primary aims are to build a virtual world that a simulated infant can explore, manipulate, and learn from; build a working prototype of a simulated infant incorporating key aspects of the BAW mental architecture; and evaluate the performance of the agent on several difficult, open-ended tasks. This architecture facilitates incorporation of key concepts from the study of natural intelligence that are infrequently used in artificial intelligence: mental models, exploratory play and chunking.
Computer science and linguistics professor James Pustejovsky is leading a Brandeis team in creating an artificial intelligence platform called Semantic Visualization of Scientific Data (SemViz) that can sort through the growing mass of published work on coronavirus and help biologists who study the disease gain insights and notice patterns and trends across research that could lead to a treatment or cure. Read the full article in BrandeisNow as well as a feature story in the Brandeis Magazine.
Cracking the Genetic Code: Two Brandeis computer scientists are using machine learning and artificial intelligence to analyze the genomes of COVID-19, other relevant corona- and avian-influenza viruses and Ebola.
Professor of computer science Pengyu Hong and assistant professor of computer science Hongfu Liu >want to identify the small and crucial bit of COVID-19's genetic code that may give rise to two of its most lethal and unique attributes. The advanced machine learning and artificial intelligence techniques the scientists are using can help sift through massive amounts of data to learn which nucleotide patterns are shared and which might be specific to COVID-19.
According to Hong and Liu, artificial intelligence may also be able to predict emerging variations in the genomes of future viruses. Read the full article on BrandeisNow.
Brandeis University was a finalist in the Northeast North American Regional Final of the International Collegiate Programming Contest (ICPC). More than 50,000 students from 3,000 universities around the world competed in the annual programming contest. Student teams were from Brandeis and 19 other colleges and universities that included RIT, Brown University, University at Buffalo, Concordia University, Harvard University, Massachusetts Institute of Technology, McGill University, Mount Allison University, Northeastern University and University of Rochester.
Congratulations to our undergraduate team, Jianfei Xue, Seeing Hu, Zhaonan Li for coming in ninth out of 20.
In the contest, each team of three students had five hours to solve a set of 10 complex, real-world problems. The top regional team will advance to the World Finals in Moscow. More information is available on the International Collegiate Programming Contest website.
WGBH announced today the Andrew W. Mellon Foundation's renewed support for WGBH with a two-year, $750,000 grant, which will enhance usability of the American Archive of Public Broadcasting (AAPB). The AAPB is a collaboration between WGBH and the Library of Congress that aims to digitize and preserve thousands of hours of broadcasts and previously inaccessible programs from the more than 60-year legacy of public radio and public television.
Over the next two years, the grant will support a two-pronged effort to make the AAPB an even more valuable resource for researchers, educators, academics and the public. The AAPB will work with Brandeis University’s Lab for Linguistics and Computation, headed by Professor James Pustejovsky, which uses machine learning and artificial intelligence to develop open-source tools and workflows, to capture detailed metadata from AAPB radio and television programs. This metadata, descriptive information about the people, places, dates and conversations in the archive, is a powerful way to improve access and discoverability of content. Read more here.
Brandeis University received a major grant to enhance the Language Application (LAPPS) Grid Project and EU’s CLARIN Platform to provide access to NLP-enabled tools to quickly analyze huge amounts of language for digital humanities to create Smart Archives. Brandeis was awarded a 16-month $673,000 grant from the Andrew W. Mellon Foundation to expand and deploy the LAPPS Grid Project which connects open-source computer programs to analyze texts from diverse sources and genres. The programs analyze any language content, determine the overall meaning and help uncover hidden relationships embedded in the data.
According to James Pustejovsky, the Project Director, the Mellon Foundation support will allow Brandeis and its collaborators from around the world to extend the range of the LAPPS Grid platform by linking it to a similarly broad and extensive one known as the European Common Language Resources and Technology Infrastructure (CLARIN). Read more here.
We are pleased to announce that Dr. Constantine Lignos has joined the Computer Science Department as an assistant professor in computational linguistics after an extensive search. If you have attended the job talks in the spring, you can probably still remember his talk on code-switching.
Dr. Lignos is currently a researcher at the University of Southern California Information Sciences Institute. Before completing a PhD in computer and information science at the University of Pennsylvania and a postdoctoral fellowship at Children's Hospital of Philadelphia. He received a BA in computer science and psychology from Yale and worked as a program manager for language technology on the Microsoft Automotive team.
Olga Papaemmanouil, associate professor of computer science, has received a prestigious AmazonResearch Award (ARA) for her proposal "Query Performance Modeling via Deep Learning" which argues for the confluence of machine learning and data management.
Professor Papaemmanouil's project focuses on leveraging deep learning methods for predicting the performance of database queries, offering flexible predictive models that automatically adapt to changes in the data distributions, workload characteristics and operational capabilities of hardware resources.
The ARA awards are granted to foster innovation and collaboration with major research institutions around the globe. The annual award offers up to $80,000 in funding to faculty members at academic institutions worldwide and $20,000 in Amazon Web Service credits to support research in a variety of artificial intelligence areas, such as computer vision, natural language processing, robotics, security and data management. This year, 82 faculty around the globe received the award; 16 of these awarded to female faculty.