Soft Q-Learning with Mutual-Information Regularization

68 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction (2019, background): “Grau-Moya et al. [17] use a learned prior over actions in the maximum entropy RL framework [26] and justify it as a regularizer based on mutual information.”
Batch Reinforcement Learning Through Continuation Method (2021, background): “Several works [27, 1, 15] studied constrained policy updates in online settings.”
Recent advances in reinforcement learning in finance (2021, background): “[97] extended the idea of Shannon’s entropy regularization to mutual-information regularization and showed its superior performance when actions have significantly different importance.”
Exploiting Hierarchy for Learning and Transfer in KL-regularized RL (2019, background): “(2018) use a pretrained policy as the default policy; other works jointly learn the policy and default policy to capture reusable behaviour from experience (Teh et al., 2017; Czarnecki et al., 2018; Galashov et al., 2019; Grau-Moya et al., 2019).”