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:20250313T164426
DTSTART;TZID=America/Detroit:20250404T143000
DTEND;TZID=America/Detroit:20250404T153000
SUMMARY:Presentation:Research Seminar
DESCRIPTION:Lara A. Boyd\, PT\, PhD\, fCAHS\, professor and distinguished university scholar at the University of British Columbia\, will speak on\, “Exploiting Neuroplasticity to enhance motor learning in healthy and damaged brains\" at this upcoming research seminar. SKB room 2200 (at the top of the north stairs\, take a sharp left\; room is in the northeast corner of the building).\n\nRSVP: https://myumi.ch/W6REd\n\nAbstract:\n\nThis talk will review recent data illustrating how neuroplastic change takes place in the human brain.  It will discuss how interventions can be applied to stimulate motor learning. Data illustrating these processes in healthy human brain will be contrasted with that from individuals who suffer from brain damage such as stroke.
UID:133850-21873620@events.umich.edu
URL:https://events.umich.edu/event/133850
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:Kinesiology
LOCATION:School of Kinesiology Building - 2200
CONTACT:
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20250203T211245
DTSTART;TZID=America/Detroit:20250404T150000
DTEND;TZID=America/Detroit:20250404T160000
SUMMARY:Lecture / Discussion:AIM Seminar:  Efficient Low-Dimensional Compression for Deep Overparameterized Learning and Fine-Tuning
DESCRIPTION:Abstract:  While overparameterization in machine learning models offers great benefits in terms of optimization and generalization\, it also leads to increased computational requirements as model sizes grow. In this work\, we demonstrate that we can reap the benefits of overparameterization without the computational burden. First\, we develop theory showing that when training the parameters of a deep linear network to fit a low-rank or wide matrix\, the gradient dynamics of each weight matrix are confined to an invariant low-dimensional subspace. This is done by carefully studying the gradient update step\, which is the product of several matrix variables\, and noticing the way low-rank structure passes from the low-rank target through the variables sequentially.  Given this invariant subspace\, we can construct and train compact\, highly compressed factorizations possessing the same benefits as their overparameterized counterparts. For language model fine-tuning\, we introduce a method called \"Deep LoRA\"\, which improves the existing low-rank adaptation (LoRA) technique. While this technique does not arise directly from our theory\, it involves only a minor modification that is surprisingly effective and of great interest for future theoretical study.\n\nContact:  Peter Miller
UID:130193-21865580@events.umich.edu
URL:https://events.umich.edu/event/130193
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:Mathematics
LOCATION:East Hall - 1084
CONTACT:
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20250303T063245
DTSTART;TZID=America/Detroit:20250404T150000
DTEND;TZID=America/Detroit:20250404T160000
SUMMARY:Careers / Jobs:Discover McKinsey's Hispanic and Latino Network
DESCRIPTION:Meet McKinsey's Hispanic and Latino Network! Hear from McKinsey consultants on life at the firm\, finding their community and more! Discover McKinsey's Hispanic and Latino Network is designed for individualswho are interested in connecting with members of McKinsey’s Hispanic and Latino Network. This is just one of our many worldwide initiatives aimed at helping individuals get to know McKinsey better.
UID:133235-21872627@events.umich.edu
URL:https://events.umich.edu/event/133235
CLASS:PUBLIC
STATUS:CONFIRMED
CATEGORIES:
LOCATION:
CONTACT:
END:VEVENT
END:VCALENDAR