CSEE Distinguished Speaker Series

Open Issues in Open World Learning

Walter J. Scheirer
Dennis O. Doughty Collegiate Professor of Engineering
University of Notre Dame

11:30-12:30 ET, Monday, March 30, 2026, UMBC ITE 459 and online

Meaningful progress has been made in open world learning (OWL), enhancing the ability of agents to detect, characterize, and incrementally learn novelty in dynamic environments. However, novelty remains a persistent challenge for agents relying on state-of-the-art learning algorithms. This talk considers the current state of OWL, drawing on insights from a recent DARPA research program on this topic. I identify open issues that impede further advancements spanning theory, design, and evaluation. In particular, I emphasize the challenges posed by dynamic scenarios that are crucial to understand for ensuring the viability of agents designed for real-world environments. The talk provides suggestions for setting a new research agenda that effectively addresses these open issues.

Dr. Scheirer's research interests within computer science include artificial intelligence, computer vision, machine learning, and digital humanities. His research has helped establish the areas of open set recognition & open world learning in computer vision.  He serves on the IEEE CS Board of Governors, CTO of the Computer Vision Foundation, and Chair Emeritus of IEEE PAMI-TC. He received his Ph.D. from the University of Colorado and his M.S. and B.A. degrees from Lehigh University. Prof. Scheirer is also a recognized cultural critic and historian, commenting on the social context of emerging technologies, such as his most recent book: A History of Fake Things on the Internet.

Using AI to Launch and Scale Your Startup

12-1 pm March 4, 2026, University Center 302, UMBC

Accessible by Design, Joy by Default: Leisure as a First-Order Accessibility Problem

Eminent Scholar Talk with Dr. Gillian Hayes, UC Irvine

12-1pm Fri., Feb. 6, ITE 459, UMBC

Refreshments provided for those that RSVP

Accessibility research has made substantial progress in education, work, and health, yet leisure—play, sport, gaming, creative expression, and shared enjoyment—often remains underemphasized or treated instrumentally. The World Leisure Organization's Charter for Leisure states, "Everyone…has the right to adequate time for rest and for the pursuit of leisure activity."  This talk explores research that understands leisure as a first-order accessibility concern rather than a secondary benefit. Taking this principle seriously invites a broader view of access as supporting dignity, self-determination, and flourishing. Drawing on a body of human-centered computing research, I examine accessibility challenges and opportunities in leisure-focused contexts: tangible technologies for blind outrigger paddlers navigating dynamic aquatic environments; digital tools that support families in managing ADHD-related self-regulation while preserving connection; and work on autism and gaming that foregrounds play on one's own terms. Across these cases, leisure settings highlight important design considerations like sensory negotiation, social coordination, challenge, and joy. By treating leisure as essential for everyone, we can think holistically about what accessibility means in everyday life.

Gillian Hayes is a Chancellor's Professor and the Kleist Professor of Informatics at the University of California, Irvine. Her research sits at the intersection of human-centered computing, accessibility, and mobile and ubiquitous computing with a particular focus on how interactive technologies can support agency, dignity, and flourishing in everyday life. Across her work, she has focused on community-engaged approaches that bring in a variety of voices to the design process who may not typically have access. Through these participatory and interdisciplinary methods, her research examines how technologies can better accommodate sensory diversity, social connection, and self-determination. When she is not doing her day job of vice provost of academic personnel or working with her students and post-docs in research, she can be found baking cakes, reading science fiction and YA dystopia, volunteering, playing with her kids, or walking her 100-pound lapdog.

UMBC's Center for Social Science Scholarship will hold a workshop series this Spring on generative AI, LLMs, and computational social science methods. Sessions will focus on familiarizing faculty and students with deploying AI and LLM models in their research and on developing a deeper understanding of the ethical, equity, and environmental consequences of these models.

February 20 | 12-1:30pm | PUP 438
Social Network Analysis: Building Web-Based Applications for Experiential Learning
Dr. Steve McDonald, Professor of Sociology, NC State University

February 27 | 12-1:30pm | PUP 438
A City in Motion: How Everyday Routines Channel and Control Crime in Baltimore
Dr. Brian Soller, Associate Professor of Sociology (SAPH), UMBC 

March 25 | 12-1:30pm | PUP 438
Geospatial Analysis
Krishna Mummadi, CS3 Graduate Assistant & GES Graduate Student

April 14 | 2-4pm | Walker Avenue, Suite 130 (Hybrid)
Foundations of Large Language Models
Dr. Josephine Namayanja, Executive Director, iHARP, UMBC
Rhoda Nankabirwa, iHARP Research Assistant and PhD Student, UMBC

April 22 | 12-1:30pm | Webex
Equity in Algorithms
Dr. Kayla Schwoerer, Assistant Professor of Public Administration, Vrije Universiteit, Amsterdam

April 29 | 12-1:30pm | Webex
Evaluating LLMs for Credible and Rigorous Social Science Research
Dr. Michael Overton, Associate Professor of Political Science and Public Administration, University of Idaho

May 1 | 12-1:30pm | PUP 438
ML Models for Causal Inference Analysis + HPC
Dr. Eric Stokan, CS3 Director and Associate Professor of Political Science, UMBC; Roy Prouty, Assistant Director for Research Computing, DoIT; Ph.D. Candidate, Computer Science, and Sai Vikas Amaraneni, iHARP Research Assistant & Ph.D. Student

Register for sessions here

Extracting Implicit Social Information from the Space Around Us

Self-Defending Ledgers: Automating Distributed Ledger Security Using Multi-Agent Reinforcement Learning and Game Theory

Md Tariqul Islam 'Pavel'
Assistant Professor of Cybersecurity
UMBC Department of Information Systems

12–1pm ET, Wednesday, December 3, 2025
UMBC ITE 459 and online via Webex

Distributed ledger technologies (DLTs) continue to face significant security challenges. While attackers constantly adapt their strategies, governance mechanisms often remain static. Our work addresses this critical gap by introducing a framework for self-defending ledgers, where nodes enforce ledger security through adaptive governance driven by multi-agent reinforcement learning (MARL) grounded in game-theoretic principles. We model DLT consensus as a repeated Bayesian game, in which participants hold probabilistic beliefs about peer behavior, allowing agents to make strategic decisions under partial observability of adversarial actions. Our framework enables nodes to model, detect, and respond to a wide range of malicious behaviors, including bribery, selfish mining, equivocation, Sybil attacks, and collusive voting, by continuously updating Bayesian trust beliefs and governance policies based on network observations. We formally prove that networks with an honest majority reach stable equilibria and provide bounds on adversarial influence. Experiments across five major protocols show that agents effectively identify attacks with high accuracy while substantially reducing adversarial success. This work demonstrates the potential of game-theoretic MARL to provide robust, self-adaptive security in varied DLT environments, paving the way for resilient and autonomous ledger governance.

Md Tariqul Islam 'Pavel' is an assistant professor of cybersecurity in UMBC's Department of Information Systems. His research centers on the security, efficiency, and fault-tolerance of distributed computing systems, with a strong emphasis on blockchain, cloud, and vehicular networks. He develops formal models, algorithms, and protocols that address critical vulnerabilities in decentralized ecosystems, spanning inter-blockchain communication, smart contract migration, and trustworthy governance. His work combines cryptography, game theory, and system design to build scalable, resilient infrastructures. He earned his PhD and MS from the University of Kentucky and BS from the University of Dhaka, Bangladesh.

On Language Models and the Dangers of an Algorithmic View From Nowhere

Dr. John Bai, University of Canterbury

2-3:00pm EST Monday, December 1, 2025, online

Large language models (LLMs) are stochastic text generators, trained using large scale, web-scrapped datasets. Uncritically accepting LLM outputs as "truthful" or "objective" risks perpetuating an algorithmic "view from nowhere", which hides the many human decisions involved in developing, training, and fine-tuning LLMs. This talk aims to demystify language modelling and support critical, de-colonial analyses of (human and artificial) knowledge systems.

Dr. John Bai works as a Senior Lecturer in the School of Psychology, Speech and Hearing at the University of Canterbury. He completed a PhD in Psychology at the University of Auckland and a Postdoc in the Department of Educational Sciences at the Carl von Ossietzky Universität Oldenburg. His research on artificial intelligence in education takes a collaborative, interdisciplinary approach to examine educators' perspectives on the possible futures of education.

This webinar is part of a series on Decentering Intercultural Competence Research and Practice, organized by the Intercultural Competence Research and Practice Working Group (the World Council on Intercultural and Global Competence) in collaboration with the UMBC Center for Intercultural Research and Practice.

Register here

Recording

Wikipedia from the World: Grounded Articles from Any Source

Alexander Martin, JHU