Optimizing Information Bottleneck in Reinforcement Learning: A Stein Variational Approach

no code implementations • 1 Jan 2021 • Pei Yingjun, Hou Xinwen, Li Jian, Lei Wang

We also show that our method achieves better performance than VIB and mutual information neural estimation (MINE), two other popular approaches to optimize the information bottleneck framework in supervised learning.

reinforcement-learning Reinforcement Learning (RL)

Learning Representations in Reinforcement Learning:An Information Bottleneck Approach

no code implementations • 12 Nov 2019 • Pei Yingjun, Hou Xinwen

We also analyze the relationship between MINE and our method, through this relationship, we theoretically derive an algorithm to optimize our IB framework without constructing the lower bound.

Representation Learning