BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.4.0.1//EN TZID:America/New_York X-WR-TIMEZONE:America/New_York BEGIN:VEVENT UID:911@fds.yale.edu DTSTART;TZID=America/New_York:20260304T113000 DTEND;TZID=America/New_York:20260304T130000 DTSTAMP:20260218T175823Z URL:https://fds.yale.edu/events/fds-colloquium-cengiz-pehlavan-harvard/ SUMMARY:FDS Colloquium: Cengiz Pehlavan (Harvard)\, "Scaling Neural Network s: Laws and Limits" DESCRIPTION:\nAbstract: Scaling up neural network models has enabled unprec edented capabilities through learning\, but we still lack the first-princi ples understanding needed to ensure their safety\, reliability\, and effic iency. I will show how tools from statistical mechanics and random matrix theory allow us to analyze neural networks in appropriate infinite-size sc aling limits\, thereby mapping the learning regimes that govern observed s caling behavior. These results account for the main features of empirical neural scaling laws\; enable transfer of near-optimal hyperparameters acro ss model sizes\, yielding significant computational benefits\; and provide a framework for understanding emergent behaviors such as in-context learn ing.\n\n\n\nSpeaker Bio: Cengiz (pronounced "Jen·ghiz") comes to Harvard SEAS from the Flatiron Institute's Center for Computational Biology (CCB)\ , where he was a a research scientist in the neuroscience group. Before CC B\, Cengiz was a postdoctoral associate at Janelia Research Campus\, and b efore that a Swartz Fellow at Harvard.\n\n\n\nCengiz received a doctorat e in physics from Brown University \;and undergraduate degrees in phys ics and electrical engineering \;from Bogazici University. He is a nat ive of Tosya\, Turkey.\n CATEGORIES:FDS Events,Colloquium LOCATION:Yale Institute for Foundations of Data Science & Webcast\, 219 Pro spect Street\, 13th Floor\, New Haven\, CT\, 06511\, United States X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=219 Prospect Street\, 13th Floor\, New Haven\, CT\, 06511\, United States;X-APPLE-RADIUS=100;X-TITLE= Yale Institute for Foundations of Data Science & Webcast:geo:0,0 END:VEVENT BEGIN:VTIMEZONE TZID:America/New_York X-LIC-LOCATION:America/New_York BEGIN:STANDARD DTSTART:20251102T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 TZNAME:EST END:STANDARD END:VTIMEZONE END:VCALENDAR