You are here

Statistics Seminar: Guang Cheng

October 18, 2018 - 11:00am

140 Nolte Center

Abstract:

We propose two novel samplers to produce high-quality samples from a given (un-normalized) probability density. The sampling is achieved by transforming a reference distribution to the target distribution with neural networks, which are trained separately by minimizing two kinds of Stein Discrepancies, and hence our method is named as Stein neural sampler. Theoretical and empirical results suggest that, compared with traditional sampling schemes, our samplers share the following three advantages: asymptotically correct; experience less convergence issue in practice; generate samples instantaneously. If time allows, some statistical interpretation for the over-fitting phenomenon arising from deep neural networks will also be mentioned. This talk is based on https://arxiv.org/pdf/1810.03545.pdf and https://arxiv.org/pdf/1810.02814.pdf