Dear All
The next Oxford Probability for Machine Learning online seminar will take place on Wednesday November 22, from 4-5 UK time.
Speaker: Krishnakumar Balasubramian (UC Davis)
Title: Optimization-based analysis of sampling algorithms
Abstract: This talk will be about recent advances in the complexity of sampling, motivated by the theory of (non-convex) optimization. The notion of first-order stationarity in sampling and a framework for understanding the iteration complexity of sampling when sampling from a non-log-concave target density will be introduced. Illustrations will be provided in the context of three different types of algorithms:
1. Langevin Monte Carlo, which is a randomized sampling algorithm
2. Regularized Stein Variational Gradient Descent, which is a deterministic sampling algorithm
3. Forward-Backward Gaussian Variational Inference, which is a (semi)-randomized sampling algorithm
Short Bio: Krishna Balasubramanian is an Associate Professor in the Department of Statistics, University of California, Davis. His research interests include Deep learning Theory, Sampling and Stochastic Optimization, as well as Geometric, Topological and Nonparametric statistics.
Join Zoom Meeting
https://zoom.us/j/97140999113?pwd=RTFzZ2VEb21PaFBEaEFBaThpdWpCZz09
Meeting ID: 971 4099 9113
Passcode: 076871
All welcome
Gesine
########################################################################
To unsubscribe from the APPLIEDPROB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=APPLIEDPROB&A=1
This message was issued to members of www.jiscmail.ac.uk/APPLIEDPROB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|