48 min
Windowed thinning and query complexity for the bouncy particle and Zigzag samplers
Computational Statistics, Computer Science and seminar by Jianfeng Lu, Duke University
Thursday 09:00 Los Angeles (GMT-7)
Recording available
Berkeley, United States
Recording
Abstract
Windowed thinning exactly simulates bouncy-particle and coordinate Zigzag sampling processes by constructing local event-rate envelopes from gradients. For strongly convex, smooth targets, the talk combines this construction with mixing estimates and event-count bounds to derive query-complexity guarantees from a Gaussian cold start, distinguishing full-gradient and coordinate-query costs.
Topics
Windowed thinningcoordinate Zigzag samplinglocal event-rate envelopesgradientssmooth targetsmixing estimatesevent-count boundsGaussian cold start