BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//UM//UM*Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Detroit
TZURL:http://tzurl.org/zoneinfo/America/Detroit
X-LIC-LOCATION:America/Detroit
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20070311T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20071104T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250120T151857
DTSTART;TZID=America/Detroit:20250320T153000
DTEND;TZID=America/Detroit:20250320T163000
SUMMARY:Workshop / Seminar:MICDE-EECS Seminar - Mikhail Belkin\, Professor\, University of California San Diego
DESCRIPTION:Bio: Mikhail Belkin is a Professor at Halicioglu Data Science Institute and Computer Science and Engineering Department at UCSD and an Amazon Scholar. Prior to that he was a Professor at the Department of Computer Science and Engineering and the Department of Statistics at the Ohio State University. He received his Ph.D. from the Department of Mathematics at the University of Chicago (advised by Partha Niyogi). His research interests are broadly in theory and applications of machine learning\, deep learning and data analysis. Some of his well-known work includes widely used Laplacian Eigenmaps\, Graph Regularization and Manifold Regularization algorithms\, which brought ideas from classical differential geometry and spectral graph theory to data science. His more recent work has been concerned with understanding remarkable mathematical and statistical phenomena observed in deep learning. The empirical evidence necessitated revisiting some of the classical concepts in statistics and optimization\, including the basic notion of over-fitting. One of his key findings has been the \"double descent\" risk curve that extends the textbook U-shaped bias-variance trade-off curve beyond the point of interpolation. His recent work focusses on understanding feature learning and over-parameterization in deep learning. Mikhail Belkin is an ACM Fellow and a recipient of a NSF Career Award and a number of best paper and other awards. He had served on the editorial boards of IEEE Proceedings on Pattern Analysis Machine Intelligence and the Journal of the Machine Learning Research. He is the editor-in-chief of SIAM Journal on Mathematics of Data Science (SIMODS).\n\nAbstract: In recent years\, transformers have become a dominant machine learning methodology.\nA key element of transformer architectures is a standard neural network (MLP). I argue that MLPs alone already exhibit many remarkable behaviors observed in modern LLMs\, including emergent phenomena. Furthermore\, despite large amounts of work\, we are still far from understanding how 2-layer MLPs learn relatively simple problems\, such as “grokking” modular arithmetic. I will discuss recent progress and argue that feature-learning kernel machines (Recursive Feature Machines) isolate some key computational aspects of modern neural architectures and are preferable to MLPs as a model for analysis of emergent phenomena.
UID:127692-21859491@events.umich.edu
URL:https://events.umich.edu/event/127692
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:Electrical And Computer Engineering,Electrical Engineering and Computer Science,Micde,Micde Seminar,Michigan Engineering,North campus
LOCATION:Electrical Engineering and Computer Science Building - 1311
CONTACT:
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20250311T151530
DTSTART;TZID=America/Detroit:20250320T153000
DTEND;TZID=America/Detroit:20250320T162000
SUMMARY:Lecture / Discussion:The Department of Astronomy 2024-2025 Colloquium Series Presents:
DESCRIPTION:\"How Long do Quasars Shine?\"\n\nLuminous quasars are believed to be the progenitors of the supermassive black holes observed ubiquitously at the centers of all massive galaxies\, but we are still in the dark about how these black holes formed. Our ignorance largely results from the fact that the expected timescale for supermassive black hole growth of 50 million years is far longer than the mere fifty years that humans have been observing quasars. A holy grail would thus be a direct measurement of quasar lifetimes\, shedding light on the physical mechanisms responsible for fueling black hole growth\, and how the back-reaction of this growth might influence how galaxies form.  I will discuss two very different experiments that allow us to construct cosmic clocks that can accurately time the duration of luminous quasar activity on timescales of kiloyears to gigayears. One exploits the clustering pattern of quasars on the sky\, which has recently been measured by JWST. The other uses observations of diffuse intergalactic gas in quasar environs. I will also touch upon how the latter can be used to constrain the reionization history of the Universe.
UID:133711-21873465@events.umich.edu
URL:https://events.umich.edu/event/133711
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:astronomy,astrophysics
LOCATION:West Hall - 411
CONTACT:
END:VEVENT
END:VCALENDAR