BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260711T111907EDT-1889OZTU6E@132.216.98.100
DTSTAMP:20260711T151907Z
DESCRIPTION:Speaker: Abhishek Halder\, Assistant Professor\, Department of 
 Applied Mathematics\, University of California at Santa Cruz\n\n \n\nAbstr
 act: \n\nMany problems in control\, statistics\, machine learning and solv
 ing partial differential equations can be cast as infinite dimensional gra
 dient flows with respect to the Wasserstein metric that appears in the the
 ory of optimal mass transport. The main idea is to construct the solution 
 of interest as the gradient descent of a suitable measure valued functiona
 l that is convex along the generalized geodesics with respect to the Wasse
 rstein metric. These connections have unfolded a rapid development across 
 several disciplines in the past two decades. Many time stepping algorithms
  are now available in the literature to numerically realize the Wasserstei
 n proximal updates\, which generalize the concept of gradient steps in the
  manifold of probability measures with finite second moments. Motivated by
  the observation that most practical problems of interest have additive ob
 jectives\, this talk will present a distributed algorithm to perform the W
 asserstein proximal updates. The proposed algorithm generalizes the finite
  dimensional Euclidean consensus ADMM algorithm to the measure valued Wass
 erstein\, and its entropy regularized Sinkhorn variants. We will point out
  how the proposed algorithm differs compared to the standard Euclidean cas
 e\, and present numerical case studies. This is joint work with Ph.D. stud
 ent Iman Nodozi at UC Santa Cruz.\n\n \n\nBio: \n\nAbhishek Halder is an A
 ssistant Professor in the Department of Applied Mathematics\, and an affil
 iated faculty in the Department of Electrical and Computer Engineering at 
 University of California\, Santa Cruz Before that he held postdoctoral pos
 itions in the Department of Mechanical and Aerospace Engineering at Univer
 sity of California\, Irvine\, and in the Department of Electrical and Comp
 uter Engineering at Texas A&M University He obtained his Bachelors and Mas
 ters from Indian Institute of Technology Kharagpur in 2008 and Ph D from T
 exas A&M University in 2014 all in Aerospace Engineering His research inte
 rests are in stochastic systems\, control and optimization with applicatio
 n focus on large scale cyber physical systems He is a co-founder of the an
 nual NorCal Control Workshop that brings together systems control research
 ers from academia and industry in the Northern California region fostering
  collaboration and professional networking.\n\n \n
DTSTART:20221104T180000Z
DTEND:20221104T190000Z
LOCATION:CA\, ZOOM
SUMMARY:A Distributed Algorithm for Wasserstein Proximal Operator Splitting
 : Theory & Applications
URL:https://www.mcgill.ca/cim/channels/event/distributed-algorithm-wasserst
 ein-proximal-operator-splitting-theory-applications-351853
END:VEVENT
END:VCALENDAR