BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//TILOS - ECPv6.18.4//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:TILOS
X-ORIGINAL-URL:https://tilos.ai
X-WR-CALDESC:Events for TILOS
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20250309T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20251102T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20260308T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20261101T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20270314T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20271107T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20280312T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20281105T090000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260515T110000
DTEND;TZID=America/Los_Angeles:20260515T120000
DTSTAMP:20260515T205603Z
CREATED:20260413T175443Z
LAST-MODIFIED:20260515T205603Z
UID:8269-1778842800-1778846400@tilos.ai
SUMMARY:Optimization for ML and AI Seminar: A non-equilibrium phase transition with broken ergodicity leads to double descent and benign overfitting in machine learning
DESCRIPTION:Nigel Goldenfeld\, UC San Diego Department of Physics and HDSI \nAbstract: The remarkable ability of modern neural networks to generalize improves with increasing network capacity\, even when the number of model parameters or effective degrees of freedom exceeds the number of training data points. This phenomenon is all the more surprising given that generalization error diverges when the number of model parameters approaches a critical value from below. Here we use dynamical mean field theory to show\, in a simple setting of linear regression\, that this so-called “double descent” behavior is the outcome of a phase transition in the stochastic field theory describing the training process. We calculate the critical exponents and scaling function of the double descent phase transition\, and show that it is marked by a breakdown of the fluctuation-dissipation theorem associated with broken ergodicity. The corresponding response function has the same functional form as the simple London model of the superconducting transition\, with the rigidity of the wave function corresponding to the neural network’s ability to generalize accurately. Our results are distinct from earlier work\, because we calculate the time-dependence specifically\, not just the equilibrium solutions. This is what enables us to identify the origin of the emergent behavior. \nWork performed with Chan Li. \n\nNigel Goldenfeld holds the Chancellor’s Distinguished Professorship in Physics at UC San Diego\, which he joined in Fall 2021 after 36 years at the University of Illinois at Urbana-Champaign (UIUC). Nigel’s research spans condensed matter theory\, the theory of living systems\, hydrodynamics and non-equilibrium statistical physics.  \nNigel received his PhD in theoretical physics from the University of Cambridge (UK) in 1982\, and from 1982-1985 was a postdoctoral fellow at the Institute for Theoretical Physics at UC Santa Barbara\, where his work on the dynamics of snowflake growth helped launch the modern theory of pattern formation in nature. He joined the condensed matter theory group at the Department of Physics at UIUC in 1985\, where his work was instrumental to the discovery of d-wave pairing in high temperature superconductors. Nigel’s interests in biology include microbial ecology\, evolution and systems biology. He was a founding member of the Institute for Genomic Biology at UIUC\, where he led the Biocomplexity Group and directed the NASA Astrobiology Institute for Universal Biology. During the COVID-19 pandemic\, he pivoted from his experience in mathematical modeling of bacteria and viruses to computational epidemiology\, advising the Governor of Illinois\, and helping devise\, set up and run the COVID saliva testing system at UIUC\, which provided ~12 hour turnaround of PCR tests to the 50\,000 people in the campus community and eventually to over 1700 schools and other institutions in Illinois and beyond. Nigel has served on the editorial boards of several journals\, including The Philosophical Transactions of the Royal Society\, Physical Biology and the International Journal of Theoretical and Applied Finance. Selected honors include: Alfred P. Sloan Foundation Fellow\, University Scholar of the University of Illinois\, the Xerox Award for research\, the A. Nordsieck award for excellence in graduate teaching and the American Physical Society’s Leo P. Kadanoff Prize 2020. Nigel is a Fellow of the American Physical Society\, a Fellow of the American Academy of Arts and Sciences\, a Fellow of the Royal Society (UK) and a Member of the US National Academy of Sciences.
URL:https://tilos.ai/event/optimization-for-ml-and-ai-seminar-a-non-equilibrium-phase-transition-with-broken-ergodicity-leads-to-double-descent-and-benign-overfitting-in-machine-learning/
LOCATION:HDSI 123 and Virtual\, 3234 Matthews Ln\, La Jolla\, CA\, 92093\, United States
CATEGORIES:TILOS Seminar Series,TILOS Sponsored Event
ATTACH;FMTTYPE=image/jpeg:https://tilos.ai/wp-content/uploads/2026/04/goldenfeld-nigel-e1776102861254.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260520T110000
DTEND;TZID=America/Los_Angeles:20260520T120000
DTSTAMP:20260520T204734Z
CREATED:20260227T004426Z
LAST-MODIFIED:20260520T204734Z
UID:8112-1779274800-1779278400@tilos.ai
SUMMARY:TILOS-HDSI Seminar: Inference-Time Algorithms: A Theoretical Lens on Tractability and Error Propagation
DESCRIPTION:Andrej Risteski\, Carnegie Mellon University \nAbstract: Modern AI systems are increasingly built by placing trained models inside larger computational loops. Inference-time algorithms are a basic instance of this idea: they use one or more trained models at test time to incorporate new information\, exploit pretrained models as priors\, and trade computational effort for accuracy\, sample quality\, or control. Examples include generator-verifier search for reasoning\, diffusion models for solving inverse problems\, and reward-guided generation. Theoretically\, this revisits a classical question from optimization and theoretical computer science: what can be done with access to an oracle? Here\, however\, the oracles are new and non-standard: they model the capabilities of large pretrained models\, making them powerful\, but also imperfect because they are learned. This combination leads to new questions about algorithm design and error propagation. \nThis talk studies two central aspects of this paradigm: computational efficiency and error propagation. The first vignette considers generator-verifier systems\, and shows how stochastic backtracking can trade additional computation for accuracy\, giving a principled version of test-time scaling even with imperfect learned oracles. The second vignette studies diffusion steering: when can we efficiently bias a pretrained diffusion model toward higher-reward samples while staying close to the original model? We show that tractability depends strongly on both the reward structure and the alignment objective\, and that simple primitives—such as sampling from linear tilts—can be surprisingly useful for handling richer reward classes. \nBased on https://arxiv.org/abs/2510.03149\, https://arxiv.org/abs/2602.16570\, https://arxiv.org/abs/2605.11361. \n\nAndrej Risteski is an Associate Professor at the Machine Learning Department in Carnegie Mellon University. Prior to that\, he was a Norbert Wiener Research Fellow jointly in the Applied Math department and IDSS at MIT. Dr. Risteski received his PhD in the Computer Science Department at Princeton University under the advisement of Sanjeev Arora. \nDr. Risteski’s research interests lie in the intersection of machine learning\, statistics\, and theoretical computer science\, spanning topics like (probabilistic) generative models\, algorithmic tools for learning and inference\, representation and self-supervised learning\, out-of-distribution generalization and applications of neural approaches to natural language processing and scientific domains. The broad goal of his research is principled and mathematical understanding of statistical and algorithmic problems arising in modern machine learning paradigms.
URL:https://tilos.ai/event/tilos-hdsi-seminar-inference-time-algorithms-a-theoretical-lens-on-tractability-and-error-propagation/
LOCATION:HDSI 123 and Virtual\, 3234 Matthews Ln\, La Jolla\, CA\, 92093\, United States
CATEGORIES:TILOS Seminar Series
ATTACH;FMTTYPE=image/png:https://tilos.ai/wp-content/uploads/2026/02/risteski-andrej-e1772152946152.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20260930T110000
DTEND;TZID=America/Los_Angeles:20260930T120000
DTSTAMP:20261002T214504Z
CREATED:20260923T171457Z
LAST-MODIFIED:20261002T214504Z
UID:8660-1790766000-1790769600@tilos.ai
SUMMARY:TILOS-SDSU Seminar: Distributionally Robust Optimization and Applications to Air Traffic Management
DESCRIPTION:Max Z. Li\, University of Michigan \nAbstract: Ground Delay Programs (GDPs) mitigate air traffic demand-capacity imbalances by holding flights on the ground at their origin airports when the destination airport’s arrival capacity is reduced\, thereby limiting costly airborne holding. A central challenge is that day-to-day demand-capacity balancing depends on accurate capacity predictions. In practice\, however\, these predictions are highly uncertain: forecast errors\, operational disruptions\, and climate-driven changes in weather severity can induce distributional shifts in capacity outcomes. As a result\, policies optimized for a single predicted distribution may perform poorly out of sample. We address this challenge by developing a distributionally robust framework for the single-airport ground holding problem (dr-SAGHP). We further propose a solution approach that integrates Kelly’s cutting plane method with the integer L-shaped method\, applicable more broadly to two-stage distributionally robust integer programs with relatively complete recourse and continuous second-stage decision variables. The approach includes a novel dual bisection and primal recovery procedure that exploits problem structure to efficiently generate the subgradients required by Kelly’s method. In computational experiments\, the proposed algorithm achieves up to two orders-of-magnitude speedups relative to directly solving the convex reformulation\, while maintaining negligible optimality gaps. Numerical results demonstrate that dr-SAGHP yields substantial out-of-sample improvements under moderate to severe distributional shifts\, enhancing the robustness and effectiveness of GDP decision-making under capacity uncertainty. \n\nMax is an Assistant Professor of Aerospace Engineering at the University of Michigan\, Ann Arbor. He also has courtesy appointments in Civil and Environmental Engineering as well as Industrial and Operations Engineering. Max received his PhD in Aerospace Engineering from the Massachusetts Institute of Technology in 2021. He received his MSE in Systems Engineering and BSE in Electrical Engineering and Mathematics\, both from the University of Pennsylvania\, in 2018. Max’s research and teaching interests include air transportation systems\, airport and airline operations\, Advanced Air Mobility\, networked systems\, as well as optimization and control. Max is the recipient of the NSF CAREER Award\, INFORMS Aviation Applications Section Best PhD Dissertation Prize\, the FAA RAISE Award\, as well as several best paper awards from national- and international-level conferences.
URL:https://tilos.ai/event/tilos-sdsu-seminar-distributionally-robust-optimization-and-applications-to-air-traffic-management/
LOCATION:Lamden Hall (LH) 247\, San Diego State University\, San Diego\, CA\, United States
CATEGORIES:TILOS Seminar Series,TILOS Sponsored Event
ATTACH;FMTTYPE=image/jpeg:https://tilos.ai/wp-content/uploads/2026/09/li-max-e1790183588208.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20261021T110000
DTEND;TZID=America/Los_Angeles:20261021T120000
DTSTAMP:20261021T224600Z
CREATED:20260930T175045Z
LAST-MODIFIED:20261021T224600Z
UID:8676-1792580400-1792584000@tilos.ai
SUMMARY:TILOS Seminar: What makes a good world model?
DESCRIPTION:Nicklas Hansen\, New York University \nAbstract: World models are quickly becoming a central ingredient in systems for decision-making. However\, several fundamental questions remain largely unresolved: what’s the right learning objective\, what’s the right training data\, and how do we evaluate the quality of a world model? In this talk\, I will present a line of work that approaches these questions from a data-centric perspective. I will first cover TD-MPC2\, a highly data-driven approach to world modeling that learns a latent predictive model optimized for decision-making rather than image generation\, and show how it enables robots to acquire new skills directly in the real world. Second\, I will turn to data and evaluation\, exploring “hallucination” in world models: producing rollouts that remain visually fluent yet are inconsistent with the true dynamics. Using a new 427-hour dataset for visual world modeling\, I will show that hallucination is\, first and foremost\, a data coverage problem\, and discuss both data- and model-centric mitigation strategies. Finally\, I will conclude the talk with an overview of open problems for the field. \n\nNicklas Hansen is a postdoctoral researcher at NYU Courant working with Lerrel Pinto and Yann LeCun. Nick’s research spans world models\, reinforcement learning\, and robot learning\, with the goal of developing generalist AI agents that learn from interaction. They completed their PhD at UC San Diego advised by Xiaolong Wang and Hao Su\, and have also spent time at NVIDIA Research\, Meta AI (FAIR)\, and UC Berkeley. Nick received their BS and MS degrees from the Technical University of Denmark\, and is a recipient of the 2024 NVIDIA Graduate Fellowship. Webpage: www.nicklashansen.com
URL:https://tilos.ai/event/tilos-seminar-what-makes-a-good-world-model/
LOCATION:Sanjay and Charlene Srivastava Multi-Purpose Room\, 3234 Matthews Ln\, La Jolla\, CA\, 92093\, United States
CATEGORIES:TILOS Seminar Series
ATTACH;FMTTYPE=image/jpeg:https://tilos.ai/wp-content/uploads/2026/09/hansen-nicklas-e1790789891971.jpeg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20270224T110000
DTEND;TZID=America/Los_Angeles:20270224T120000
DTSTAMP:20261026T213038Z
CREATED:20261026T212033Z
LAST-MODIFIED:20261026T213038Z
UID:9067-1803466800-1803470400@tilos.ai
SUMMARY:TILOS Seminar with Thomas Serre (Brown)
DESCRIPTION:Thomas Serre\, Brown University \nAbstract: TBA \n\nDr. Serre is the Thomas J. Watson\, Sr. Professor of Science and Professor of Cognitive and Psychological Sciences and Computer Science at Brown University. He received his Ph.D. in Neuroscience from MIT in 2006 and his M.Sc. in Electrical Engineering and Computer Science from Télécom Bretagne in France in 2000. His research spans computational neuroscience\, NeuroAI\, and computer vision. He studies the computational principles of biological vision\, including the roles of recurrent and feedback processing and neural dynamics\, and develops brain-inspired models and benchmarks that compare human and machine vision. His work has been featured by the BBC\, The Economist\, New Scientist\, and Scientific American. \nDr. Serre serves as the Faculty Director of the Center for Computation and Visualization and the Associate Director of the Center for Computational Brain Science. He is also an affiliate of the Carney Institute for Brain Science\, the Center for Theoretical Physics and Innovation\, and the Data Science Institute at Brown University. He holds an International Chair in Artificial Intelligence at the Artificial and Natural Intelligence Toulouse Institute in France and is an elected Fellow of the European Laboratory for Learning and Intelligent Systems (ELLIS). \nHe has served as an area chair and senior program committee member for machine learning and computer vision conferences\, including AAAI\, CVPR\, ICML\, ICLR\, and NeurIPS. He is a Senior Editor of Minds\, Machines\, and Brains (MIT Press) and served as Deputy Editor and Neuroscience Section Editor of PLOS Computational Biology from 2021 to 2026. \nHis awards include an NSF Early Career Award\, a DARPA Young Faculty Award\, a DARPA Director’s Award\, and Brown’s Mid-Career Research Achievement Award (2024). With his team\, he received the 2021 IEEE PAMI Helmholtz Prize and the 2022 IEEE PAMI Mark Everingham Prize for work on human action recognition. \nZoom: https://bit.ly/TILOS-Seminars
URL:https://tilos.ai/event/tilos-seminar-with-thomas-serre-brown/
LOCATION:Sanjay and Charlene Srivastava Multi-Purpose Room\, 3234 Matthews Ln\, La Jolla\, CA\, 92093\, United States
CATEGORIES:TILOS Seminar Series
ATTACH;FMTTYPE=image/jpeg:https://tilos.ai/wp-content/uploads/2026/10/serre-thomas-e1793049550526.jpeg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20270310T110000
DTEND;TZID=America/Los_Angeles:20270310T120000
DTSTAMP:20261026T213015Z
CREATED:20261026T212419Z
LAST-MODIFIED:20261026T213015Z
UID:9073-1804676400-1804680000@tilos.ai
SUMMARY:TILOS Seminar with Atul Prakash (Michigan)
DESCRIPTION:Atul Prakash\, University of Michigan \nAbstract: TBA \n\nAtul Prakash is a Professor and Chair of Computer Science and Engineering at the University of Michigan. He earned his Ph.D. from the University of California\, Berkeley\, and joined the Michigan faculty in 1989. His research focuses on the efficiency\, robustness\, and security of machine learning and computer systems\, including large language models. His group showed that a few stickers on a stop sign can fool state-of-the-art image classifiers. That work helped inspire DARPA’s GARD program\, and the sign was exhibited at the London Science Museum. His honors include a Distinguished Practical Paper Award from the IEEE Symposium on Security and Privacy. At Michigan\, he also led the creation of the undergraduate Data Science program. \nZoom: https://bit.ly/TILOS-Seminars
URL:https://tilos.ai/event/tilos-seminar-with-atul-prakash-michigan/
LOCATION:Sanjay and Charlene Srivastava Multi-Purpose Room\, 3234 Matthews Ln\, La Jolla\, CA\, 92093\, United States
CATEGORIES:TILOS Seminar Series
ATTACH;FMTTYPE=image/jpeg:https://tilos.ai/wp-content/uploads/2026/10/prakash-atul-e1793049578948.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Los_Angeles:20270428T110000
DTEND;TZID=America/Los_Angeles:20270428T120000
DTSTAMP:20261026T212948Z
CREATED:20261026T212730Z
LAST-MODIFIED:20261026T212948Z
UID:9076-1808910000-1808913600@tilos.ai
SUMMARY:TILOS Seminar with Stella Yu (Michigan)
DESCRIPTION:Stella Yu\, University of Michigan \nAbstract: TBA \n\nStella Yu received her Ph.D. from Carnegie Mellon University\, where she studied robotics at the Robotics Institute and vision science at the Center for the Neural Basis of Cognition. Before joining the University of Michigan in Fall 2022 as a Full Professor of Electrical Engineering and Computer Science\, she was Director of the Vision Group at the International Computer Science Institute\, a Senior Fellow at the Berkeley Institute for Data Science\, and an Adjunct Professor in Computer Science\, Vision Science\, and Cognitive and Brain Sciences at UC Berkeley. Dr. Yu is a recipient of the NSF CAREER Award and the Clare Boothe Luce Professorship.  Her research explores how embodied intelligence can emerge from natural data and rich interaction\, enabling intelligent systems to perceive\, act\, and adapt with greater autonomy and minimal external supervision.  Her long-term vision is to develop robots that learn like humans\, collaborate with and assist people\, and unlock new possibilities for human work and life. \nZoom: https://bit.ly/TILOS-Seminars
URL:https://tilos.ai/event/tilos-seminar-with-stella-yu-michigan/
LOCATION:Sanjay and Charlene Srivastava Multi-Purpose Room\, 3234 Matthews Ln\, La Jolla\, CA\, 92093\, United States
CATEGORIES:TILOS Seminar Series
ATTACH;FMTTYPE=image/jpeg:https://tilos.ai/wp-content/uploads/2026/10/yu-stella-e1793049563723.jpg
END:VEVENT
END:VCALENDAR