February 2026 - talks - lists.cs.princeton.edu

CS Colloquium Speaker: Stefano Soatto, Tues. Feb 24 at 3:20pm
by Emily C. Lawrence 17 Mar '26

17 Mar '26

CS Colloquium Speaker: Stefano Soatto from UCLA & Amazon Date & Time: Tuesday, February 24, 2026 - 3:20pm Location: CS 105 Host: Jia Deng Webinar registration: [ https://princeton.zoom.us/webinar/register/WN_geQ0rhfcQR-fwkdQGAo7Ug | https://princeton.zoom.us/webinar/register/WN_geQ0rhfcQR-fwkdQGAo7Ug ] Title: AI Agents as Universal Task Solvers: Time, Information, and Intelligence in LLMs viewed as Maximalistic Models of Computation Abstract: Scaling laws predict that AI agents will steadily improve and eventually exceed human performance across a wide range of tasks. Yet at the limit of these scaling laws lies a form of inference that involves no intelligence at all: with increasing compute and memory, a model can brute-force any verifiable task without learning anything from past experience. Universally optimal inference, pioneered by Solomonoff and Levin, requires no insight — only exhaustive search. This raises a basic question: if scaling alone does not foster intelligence, what does? And if performance on downstream tasks is insufficient to measure intelligence, what is? In this talk, I will point to the critical role of time in both analyzing and fostering the emergent reasoning behavior of AI agents. Building on insights that Solomonoff sketched in 1985 but that remained theoretical curiosities for decades, I will show that the value of learning is measured not by a reduction in uncertainty — the core of inductive learning and generalization — but by a reduction in the time needed to solve new tasks. A key result is that data can make a universal solver exponentially faster, with the speed-up tightly characterized by a single quantity: the algorithmic mutual information between past experience and the solution to unforeseen tasks. Connecting these ideas to modern AI requires rethinking what computation means for systems powered by large language models. Unlike minimalistic models of computation such as Turing Machines, LLMs are stochastic dynamical systems whose computational elements — context, weights, activations, chain-of-thought — do not resemble a "program" in the ordinary sense. I will show that LLMs are instead maximalistic models of computation: universal, like Turing Machines, but operating through entirely different and in many ways antithetical mechanisms. Programming such systems can be achieved through two-level control strategies — open-loop planning and closed-loop feedback — in abstract space, a framework we have recently released Strands Agents open-source library (www.strandsagents.com). Once time is properly accounted for, scaling laws reveal an inversion : beyond a critical point, increasing resources improve benchmark accuracy while diminishing conceptual depth— a savant regime in which models improve while learning less. I will discuss what this means for how we build, evaluate, and scale AI agents. Bio: Stefano Soatto is a Vice President at AWS Agentic AI, and a Professor of Computer Science at UCLA. He received his PhD in Control and Dynamical Systems from the California Institute of Technology, his D.Ing from the University of Padova, Italy and was a postdoctoral scholar at Harvard University. He is a Fellow of the ACM and of the IEEE.

1 4

ECE Seminar: Weijia Shi, Monday, March 2
by Emily C. Lawrence 27 Feb '26

27 Feb '26

ECE SEMINAR Speaker: Weijia Shi, University of Washington Day: Monday, March 2, 2026 Time: 4:30 PM Location: B205 Engineering Quadrangle Host: Tom Silver Title: Breaking the Language Model Monolith Abstract: Language models (LMs) are typically monolithic: a single model storing all knowledge and serving every use case. This design presents significant challenges; they often generate factually incorrect statements, require costly retraining to add or remove information, and face serious privacy and copyright issues. In this talk, I will discuss how to break this monolith by introducing modular architectures and training algorithms that separate capabilities across composable components. I'll cover two forms of modularity: (1) External modularity, which augments LMs with external tools like retrievers to improve factuality and reasoning; and (2) internal modularity, which builds inherently modular LMs from decentrally trained components to enable flexible composition and an unprecedented level of control. Bio: Weijia Shi is a Ph.D. candidate at the University of Washington, where she is advised by Luke Zettlemoyer and Noah Smith. Her research focuses on developing augmented and modular architectures and training algorithms that enable language models to be more controllable, collaboratively developed, and factual. She received an Outstanding Paper Award at ACL 2024 and was recognized as Rising Stars in Machine Learning in 2023 and in Data Science in 2024.

1 0

ECE Seminar: Yu Sun TODAY at 4:30
by Emily C. Lawrence 26 Feb '26

26 Feb '26

ECE SEMINAR Speaker: Yu Sun, Stanford University Title: Test-Time Training Day: Thursday, February 26, 2026 Time: 4:30 PM Location: B205 Engineering Quadrangle Host: Dhruv Shah Abstract: Most AI models are trained only before the test instances arrive and then fixed during deployment, even though making good predictions on test instances is the ultimate goal of training. What if we continue to train a model after each test instance arrives? In this talk, we discuss how this conceptual framework, known as test-time training, leads to long-term memory that scales differently with context length, and enables AI to discover new results on open scientific problems. Bio: Yu Sun is a postdoc at Stanford University and a researcher at NVIDIA. His research focuses on continual learning, specifically a conceptual framework known as test-time training, where each test instance defines its own learning problem. Yu obtained his PhD in EECS from UC Berkeley and BS in CS from Cornell University.

1 0

Computational Math: IDeAS seminar, Thurs at 2pm
by Emily C. Lawrence 25 Feb '26

25 Feb '26

COMPUTATIONAL MATH – IDeAS SEMINAR Recurring weekly series · Thursdays, 2:00 – 3:00 PM · Jadwin 111 This week's talk: Speaker: Ethan Nicholas Epperly (UC Berkeley) Title: Sparser, better, faster, stronger: Emerging approaches in the theory and practice of randomized dimensionality reduction. Date: Thursday, February 26, 2026 Time: 2:00 PM – 3:00 PM Room: Jadwin 111 Abstract: For approaching thirty years, randomized dimensionality reduction techniques have been a core tool in the theory and practice of computation. But despite this robust history, basic questions remain hotly debated: Which dimensionality reduction map should be used? How can the map be adapted to structure in the problem? What is the right theoretical approach to analyzing randomized dimensionality reduction? This talk presents a new approach to the theory of randomized dimensionality reduction under which the core feature of a good dimensionality map is injectivity: It is fine if the map stretches things out a bit as long as it does not annihilate any element of the data space. This new approach yields new analysis of sparse and tensor-structured dimensionality reduction maps that comes closer to describing how these maps behave in practice. This talk is designed for a general audience and assumes no prior familiarity with randomized dimensionality reduction. About the speaker: Ethan is a Miller Research Fellow in the Mathematics Department at UC Berkeley. His research is focused on designing computational techniques for solving large-scale computational problems in machine learning, quantum information, and scientific computing, with a focus on randomized algorithms. He received his PhD in Applied and Computational Mathematics from Caltech. He has been recognized for his work with finalist status for the Hertz Foundation Fellowship, Caltech's Thomas A. Tisch Prize for Graduate Teaching in CMS and W. P. Carey & Co. Prize in Applied Mathematics, and the SIAM Student Paper Prize. His PhD research was supported by a Department of Energy Computational Science Graduate Fellowship. Up next: Thursday, March 5 – Per-Gunnar Martinsson (The University of Texas). Same time, same place. The IDeAS Seminar series is hosted by Liza Rebrova (ORFE), Marc Aurèle Gilles and Jorge Garza Vargas (MATH).

1 0

[Robotics Seminar] TOMORROW FEB 20 @ 11AM Naira Hovakimyan
by Emily C. Lawrence 19 Feb '26

19 Feb '26

Join us TOMORROW (02/20) for our first Robotics Seminar of the semester. Please note the change of location to CS 105. Speaker: [ https://mechse.illinois.edu/people/profile/nhovakim | Naira Hovakimyan ] Date: Feb 20th, 2026 Time: 11:00 AM EST Location: CS 105 Event page: [ https://robotics.princeton.edu/seminars | https://robotics.princeton.edu/seminars ] The seminar will be followed by a lunch with the speaker in the F-Wing space. We are looking forward to seeing you there! Best wishes, Robotics Seminar Committee Title: Ontological Robustness for Certification of Layered Autonomy Architectures Abstract: Learning-based control paradigms have seen many success stories with autonomous systems in recent years. A typical architecture in these systems involves layers for perception, planning and control, wherein each of these layers uses different tools and metrics for assessing robustness and performance. For example, the planners -- that use vision-based sensors to update the navigation and motion planning -- operate largely relying on distributionally robust stochastic optimal control, whereas the low-level controller can be a deterministic controller with its conventional gain and phase (time-delay) margin. We present a new analysis framework for addressing this ontology challenge inherent to autonomous systems. We derive distributional robustness guarantees for deterministic L1 adaptive controllers that can be used by any stochastic planner without facing a language barrier. The combined planner-controller framework can serve as foundation for development of certificates for V&V of learning-enabled systems. An overview of different projects at our lab that build upon this framework will be demonstrated to show different applications. Bio: Naira Hovakimyan received her MS degree in Applied Mathematics from Yerevan State University in Armenia. She got her Ph.D. in Physics and Mathematics from the Institute of Applied Mathematics of Russian Academy of Sciences in Moscow. She is currently W. Grafton and Lillian B. Wilkins Professor of Mechanical Science and Engineering and the Director of AVIATE Center of UIUC. She has co-authored two books, eleven patents and more than 500 refereed publications. She is the 2011 recipient of AIAA Mechanics and Control of Flight Award , the 2015 recipient of SWE Achievement Award, the 2017 recipient of IEEE CSS Award for Technical Excellence in Aerospace Controls , and the 2019 recipient of AIAA Pendray Aerospace Literature Award . In 2014 she was awarded the Humboldt prize for her lifetime achievements. In 2015 and 2023 she was awarded the UIUC Engineering Council Award for Excellence in Advising . In 2024 she was recognized as the winner of the College Award for Excellence in Translational Research , and in 2025 she was recognized for Excellence in Graduate Student Mentoring . She is Fellow of AIAA, IEEE, ASME, IFAC, and senior member of National Academy of Inventors. She has been named a Distinguished Lecturer for IEEE CSS for 2026-2028. She is a co-founder and chief scientist of Intelinair. Her work in robotics for elderly care was featured in the New York Times, on Fox TV, CNBC, and her recent NASA ULI award on flying cars led her to a live interview on Cheddar Innovates and many other media platforms. Her research interests are in control and optimization, autonomous systems, machine learning, neural networks, game theory, and their applications in aerospace, robotics, mechanical, agricultural, electrical, petroleum, biomedical engineering, and elderly care.

1 0

[Robotics Seminar] FRIDAY FEB 20 @ 11AM Naira Hovakimyan
by Emily C. Lawrence 13 Feb '26

13 Feb '26

Join us for our first Robotics Seminar of the semester next Friday (02/20) . Please note the change of location to CS 105. Speaker: [ https://mechse.illinois.edu/people/profile/nhovakim | Naira Hovakimyan ] Date: Feb 20th, 2026 Time: 11:00 AM EST Location: CS 105 Event page: [ https://robotics.princeton.edu/seminars | https://robotics.princeton.edu/seminars ] The seminar will be followed by a lunch with the speaker in the F-Wing space. We are looking forward to seeing you there! Best wishes, Robotics Seminar Committee Title: Ontological Robustness for Certification of Layered Autonomy Architectures Abstract: Learning-based control paradigms have seen many success stories with autonomous systems in recent years. A typical architecture in these systems involves layers for perception, planning and control, wherein each of these layers uses different tools and metrics for assessing robustness and performance. For example, the planners -- that use vision-based sensors to update the navigation and motion planning -- operate largely relying on distributionally robust stochastic optimal control, whereas the low-level controller can be a deterministic controller with its conventional gain and phase (time-delay) margin. We present a new analysis framework for addressing this ontology challenge inherent to autonomous systems. We derive distributional robustness guarantees for deterministic L1 adaptive controllers that can be used by any stochastic planner without facing a language barrier. The combined planner-controller framework can serve as foundation for development of certificates for V&V of learning-enabled systems. An overview of different projects at our lab that build upon this framework will be demonstrated to show different applications. Bio: Naira Hovakimyan received her MS degree in Applied Mathematics from Yerevan State University in Armenia. She got her Ph.D. in Physics and Mathematics from the Institute of Applied Mathematics of Russian Academy of Sciences in Moscow. She is currently W. Grafton and Lillian B. Wilkins Professor of Mechanical Science and Engineering and the Director of AVIATE Center of UIUC. She has co-authored two books, eleven patents and more than 500 refereed publications. She is the 2011 recipient of AIAA Mechanics and Control of Flight Award , the 2015 recipient of SWE Achievement Award, the 2017 recipient of IEEE CSS Award for Technical Excellence in Aerospace Controls , and the 2019 recipient of AIAA Pendray Aerospace Literature Award . In 2014 she was awarded the Humboldt prize for her lifetime achievements. In 2015 and 2023 she was awarded the UIUC Engineering Council Award for Excellence in Advising . In 2024 she was recognized as the winner of the College Award for Excellence in Translational Research , and in 2025 she was recognized for Excellence in Graduate Student Mentoring . She is Fellow of AIAA, IEEE, ASME, IFAC, and senior member of National Academy of Inventors. She has been named a Distinguished Lecturer for IEEE CSS for 2026-2028. She is a co-founder and chief scientist of Intelinair. Her work in robotics for elderly care was featured in the New York Times, on Fox TV, CNBC, and her recent NASA ULI award on flying cars led her to a live interview on Cheddar Innovates and many other media platforms. Her research interests are in control and optimization, autonomous systems, machine learning, neural networks, game theory, and their applications in aerospace, robotics, mechanical, agricultural, electrical, petroleum, biomedical engineering, and elderly care.

1 0

ECE Seminar: Max Fishelson, MIT on Monday, Feb 16
by Emily C. Lawrence 12 Feb '26

12 Feb '26

ECE SEMINAR Speaker: Max Fishelson, Massachusetts Institute of Technology Title: Theoretical Foundations for Multi-Agent Learning Day: Monday, February 16, 2026 Time: 4:30 PM Location: B205 Engineering Quadrangle Host: Pramod Viswanath Abstract: As artificial intelligence becomes ubiquitous in complex, interactive systems, we need models that perform reliably under dynamic conditions. AI systems are moving beyond passive perception to active participation: controlling autonomous vehicles in mixed traffic, participating in markets, and managing decentralized networks. In these roles, an agent's decisions reshape the environment and invite strategic responses from others. This feedback loop invalidates a core assumption of classical machine learning: that data is independent and identically distributed. We require a new foundation for learning that is inherently robust to these adaptive, multi-agent dynamics. In this talk, I will demonstrate how we can approach this challenge using regret minimization and calibration. Instead of measuring performance against a static distribution, these measures allow us to derive provable guarantees against any sequence of outcomes. An algorithm with low regret effectively “learns to play” optimally, even against strategic counter-parties. Calibration guarantees that we can distill trustworthy forecasts from dynamic data. I will discuss my work pushing the boundaries of what is possible in this domain, resolving long-standing open questions on regret minimization in games, the scalability of algorithms for strong notions of regret, and the fundamental limits of trustworthy forecasting. Together, these results provide an algorithmic pathway toward AI systems that remain robust and reliable even in complex, interactive environments. Bio: Max Fishelson is a final-year PhD student in the Theory of Computation group at MIT CSAIL, advised by Constantinos Daskalakis. His research focuses on the theoretical foundations of machine learning, specifically at the intersection of learning theory and algorithmic game theory. He works on designing algorithms with provable guarantees for decision-making in adversarial and multi-agent environments.

1 0

William Pierson Field Lecture: Auditing Automated Speech Recognition Systems for Disparities - Tues, Feb 17 at 1:30pm
by Emily C. Lawrence 10 Feb '26

10 Feb '26

William Pierson Field Lecture Date: Tuesday, Feb 17th Time: 1:30 – 2:30pm Location: Louis A. S. A71 Speaker: Allison Koenecke, Cornell Tech Host: Lydia Liu Title: [ https://www.cs.princeton.edu/events/auditing-automated-speech-recognition-s… | Auditing Automated Speech Recognition Systems for Disparities ] Bio: Allison Koenecke is an Assistant Professor of Information Science at Cornell Tech. Her research on algorithmic fairness applies computational methods, such as machine learning and causal inference, to study societal inequities in domains from online services to public health. Koenecke is regularly quoted as an expert on disparities in automated speech-to-text systems. She previously held a postdoctoral researcher role at Microsoft Research and received her PhD from Stanford's Institute for Computational and Mathematical Engineering. She is the recipient of several NSF grants and a Cornell CIS DEIB Faculty of the Year Award, and has been honored as a Sloan Fellow in Computer Science and Forbes 30 Under 30 lister in Science.

1 0

CS Distinguished Colloquium Speaker Jeff Dean: Tues. Feb 10 at 4:30
by Emily C. Lawrence 10 Feb '26

10 Feb '26

Distinguished Colloquium Speaker: Jeff Dean from Google Date & Time: Tuesday, February 10 - 4:30pm Location: CS 104 Host: Kai Li Title: Important Trends in AI: How Did We Get Here, What Can We Do Now, and Where are We Headed? Abstract: In this talk I'll highlight some of the key developments in artificial intelligence over the past 15 years, which serve as key ingredients in today's most advanced AI models. This will touch on development of new model architectures, advances in large-scale distributed training, ML accelerators like TPUs, and development of algorithms to improve training and serving efficiency. I'll discuss the Gemini effort at Google, both in terms of the capabilities of Gemini models but also in how we go about organizing a large-scale research and engineering effort like Gemini. Finally, I'll give a sense of the capabilities of today's models, and also highlight some areas that will advance rapidly in the next few years. Bio: Jeff Dean is the Chief Scientist for Google Research and Google DeepMind, and currently co-leads the Gemini project. He has done research in large-scale information retrieval, operating systems, distributed systems, compilers and in the field of AI, has done work on ML accelerators, vision models, low-level software and frameworks for machine learning, sparse model architectures, algorithms like distillation and neural architecture search, large language and multimodal models, and applications of machine learning to areas like ASIC design, healthcare, and translation. He is a recipient of the ACM Prize in Computing, and the IEEE John von Neumann medal, and a Fellow of the ACM, and a member of the US National Academy of Engineering, and the American Academy of Arts and Sciences. *This talk will not be live streamed.

1 1

PPH seminar: Sneha Goenka-Biology Needs Computer Architects: Keeping Up with Genomic-scale Data, Fri. Feb 6
by Emily C. Lawrence 04 Feb '26

04 Feb '26

Our next talk in the PPH seminar series is on Friday, February 6, at 12 pm , at 252 Nassau Street. Sneha Goenka will present, "Biology Needs Computer Architects: Keeping Up with Genomic-scale Data. " Professor Sneha Goenka is an Assistant Professor in the ECE department at Princeton University. Lunch will be provided. Abstract: The landscape of computing has undergone a significant transformation with the death of Dennard scaling and the slowing of Moore’s law: applications now drive innovations in computer systems architecture. At the same time, the advent of high throughput, low-cost sequencing technology has revolutionized genomics. The massive amount of data generated in genomics has uncovered acute compute challenges in biological inference due to limitations of software on traditional multi-core systems. To address this challenge, my research employs a hardware-software-algorithm co-design approach to significantly improve computational performance in key areas of comparative and clinical genomics. In this talk, I will present systems that accelerate pipelines in both these domains of genomics. First, I will present Darwin-WGA (FPGA/ASIC) and SegAlign (GPU) for cross-species whole genome alignment where co-design has yielded orders of magnitude increase in speed. Additionally, there are gains in accuracy while modifying the algorithm to improve the underlying hardware implementation. Next, I will talk about the ultra-rapid nanopore whole genome sequencing pipeline that can deliver a genetic diagnosis in under 8 hours, making it the fastest pipeline to date. The scalable, cloud-based distributed infrastructure overcomes system bottlenecks to enable near real-time computation and improved variant identification. This pipeline has been deployed in critical care units in Stanford hospitals and applied to 25 patients. Finally, I will share my vision and our current work on the continued advancements in computer systems applied to emerging fields such as pangenomics, single cell genomics, cancer and long read clinical sequencing. Upcoming talks include the following - please check our [ https://us.list-manage.com/fbV_-3DQ__6?e=691aac62ef&c2id=f584458f8dee0f3191… | website ] for more information and updates! Monday, Feb 9, at 10:00 am **NOTE THE SPECIAL DATE AND TIME** Stephany Duda (Vanderbilt) [ https://us.list-manage.com/x7_GHnGq-1c?e=691aac62ef&c2id=f584458f8dee0f3191… | Beyond Data Collection: REDCap as a Platform for Research Collaboration ] Friday, Feb 27, Aayush Visaria (Rutgers Robert Wood Johnson Medical School) ** AI and Precision Health series ** AI and Precision Health: [ https://us.list-manage.com/6_L2WELiCM8?e=691aac62ef&c2id=f584458f8dee0f3191… | Beyond Fat Mass: How Adiposity, Oxidative Stress, and Bilirubin Shape Metabolic Risk ] Friday, March 6, Florencia Torche (Princeton): [ https://us.list-manage.com/JR1vCff5h6C?e=691aac62ef&c2id=f584458f8dee0f3191… | Early Life Exposures and Social Inequality ] Friday, March 20, Ricky Bloomfield (Oura) ** AI and Precision Health series ** [ https://us.list-manage.com/_rvWqr__Bca?e=691aac62ef&c2id=f584458f8dee0f3191… | AI and Precision Health: From Episodic Care to Continuous Health: How Wearables and AI Are Reshaping Preventive Care ] Friday, April 3, David Pagliaccio (Columbia) Friday, April 10, René Vidal (UPenn) ** AI and Precision Health series ** Friday, April 24, Tom Silver (Princeton) Friday, May 8, Jessica Hamilton (Rutgers) [ https://us.list-manage.com/QW-l0J1AO1x?e=691aac62ef&c2id=f584458f8dee0f3191… | View all upcoming PPH events ] . Please note that getting to the seminar space at 252 Nassau Street currently requires that you climb a set of stairs. If an accommodation is needed, please contact PPH in advance at: [ mailto:princetonPPH@princeton.edu | princetonPPH(a)princeton.edu ] Thank you, Princeton Precision Health

1 1