pith. sign in

arxiv: 1908.04970 · v2 · pith:SQ3LR5H2new · submitted 2019-08-14 · 💻 cs.LG · stat.ML

Thompson Sampling with Approximate Inference

classification 💻 cs.LG stat.ML
keywords alphainferencesamplingthompsonapproximateconstanterroreven
0
0 comments X
read the original abstract

We study the effects of approximate inference on the performance of Thompson sampling in the $k$-armed bandit problems. Thompson sampling is a successful algorithm for online decision-making but requires posterior inference, which often must be approximated in practice. We show that even small constant inference error (in $\alpha$-divergence) can lead to poor performance (linear regret) due to under-exploration (for $\alpha<1$) or over-exploration (for $\alpha>0$) by the approximation. While for $\alpha > 0$ this is unavoidable, for $\alpha \leq 0$ the regret can be improved by adding a small amount of forced exploration even when the inference error is a large constant.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.