第61回統計的機械学習セミナー / The 61st Statistical Machine Learning Seminar
- 【日時】
- 2024年7月4日(木) 16:00 - 17:30
参加無料 / Admission Free
- 【場所】
- 統計数理研究所・D棟3階セミナー室5 (ハイブリッド)
オンライン参加を希望される場合は、以下の google form に登録し、Zoom情報をお受け取りください．
https://forms.gle/RVB9aYh2GSQfTDjf8
(現地参加の場合は登録不要です)
【Speaker】
Heishiro Kanagawa (Newcastle University)
【Title】
Reinforcement Learning for Adaptive MCMC
【Abstract】
An informal observation, made by several authors, is that the adaptive design of a Markov transition kernel has the flavour of a reinforcement learning task. Yet, to-date it has remained unclear how to actually exploit modern reinforcement learning technologies for adaptive MCMC. The aim of this work is to set out a general framework, called Reinforcement Learning Metropolis--Hastings, that is theoretically supported and empirically validated. Our principal focus is on learning fast-mixing Metropolis--Hastings transition kernels, which we cast as deterministic policies and optimise via a policy gradient. Control of the learning rate provably ensures conditions for ergodicity are satisfied. The methodology is used to construct a gradient-free sampler that out-performs a popular gradient-free adaptive Metropolis--Hastings algorithm on ≈90% of tasks in the PosteriorDB benchmark.
【主催】
統計数理研究所 先端データサイエンス研究系 統計的機械学習研究センター
【連絡先】
福水健次
E-mail: fukumizu
ism.ac.jp
