mlsanityData to 14 Oct 2026

Enforcing KL Regularization in General Tsallis Entropy Reinforcement Learning via Advantage Learning

arXiv 2205.07885 ↗First posted 16 May 2022cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Maximum Tsallis entropy (MTE) framework in reinforcement learning has gained popularity recently by virtue of its flexible modeling choices including the widely used Shannon entropy and sparse entropy. However, non-Shannon entropies suffer from approximation error and subsequent underperformance either due to its sensitivity or the lack of closed-form policy expression. To improve the tradeoff between flexibility and empirical performance, we propose to strengthen their error-robustness by enforcing implicit Kullback-Leibler (KL) regularization in MTE motivated by Munchausen DQN (MDQN). We do so by drawing connection between MDQN and advantage learning, by which MDQN is shown to fail on generalizing to the MTE framework. The proposed method Tsallis Advantage Learning (TAL) is verified on extensive experiments to not only significantly improve upon Tsallis-DQN for various non-closed-form Tsallis entropies, but also exhibits comparable performance to state-of-the-art maximum Shannon entropy algorithms.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 33 references that are in the map. First posted on arXiv 16 May 2022.

Builds on

15 of 33 references.

References · 15

  1. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  2. 2018Addressing Function Approximation Error in Actor-Critic Methodscited by 2,417
  3. 2018Learning Dexterous In-Hand Manipulationcited by 687
  4. 2017Distributional Reinforcement Learning with Quantile Regressioncited by 474
  5. 2016From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classificationcited by 341
  6. 2019A Theory of Regularized Markov Decision Processescited by 286
  7. 2018Understanding the impact of entropy on policy optimizationcited by 176
  8. 2019Diagnosing Bottlenecks in Deep Q-learning Algorithmscited by 73
  9. 2020Munchausen Reinforcement Learningcited by 69
  10. 2015Increasing the Action Gap: New Operators for Reinforcement Learningcited by 49
  11. 2017Sparse Markov Decision Processes with Causal Sparse Tsallis Entropy Regularization for Reinforcement Learningcited by 37
  12. 2020Leverage the Average: an Analysis of KL Regularization in RLcited by 33
3 more
  1. 2018Path Consistency Learning in Tsallis Entropy Regularized MDPscited by 22
  2. 2020Self-Imitation Advantage Learningcited by 15
  3. 2018Effective Exploration for Deep Reinforcement Learning via Bootstrapped Q-Ensembles under Tsallis Entropy Regularizationcited by 8

Built on by

Not cited yet.