A Short Survey on Probabilistic Reinforcement Learning

2019/01/21 by Reazul Hasan Russel, Russel, Reazul Hasan
Computer Science · Decision Sciences · Mathematics · #Advanced Bandit Algorithms Research #Data Stream Mining Techniques #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Reinforcement Learning in Robotics #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.1901.07010

7 pages, originally written as a literature survey for PhD candidacy exam

arxiv created 2019/01/21 · openalex publication_date 2019/01/21 · arxiv updated 2019/01/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

A reinforcement learning agent tries to maximize its cumulative payoff by interacting in an unknown environment. It is important for the agent to explore suboptimal actions as well as to pick actions with highest known rewards. Yet, in sensitive domains, collecting more data with exploration is not always possible, but it is important to find a policy with a certain performance guaranty. In this paper, we present a brief survey of methods available in the literature for balancing exploration-exploitation trade off and computing robust solutions from fixed samples in reinforcement learning.

Citations

• Sample Complexity of Episodic Fixed-Horizon Reinforcement Learning
• Distributionally Robust Markov Decision Processes
• Robust Markov Decision Processes
• Robust Dynamic Programming
• Safe Policy Improvement by Minimizing Robust Baseline Regret
• #Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
• Deep Exploration via Randomized Value Functions
• Why is Posterior Sampling Better than Optimism for Reinforcement Learning?

Related

• Making Sense of Reinforcement Learning and Probabilistic Inference
• Beyond Optimism: Exploration With Partially Observable Rewards
• Reinforcement Learning with Probabilistically Complete Exploration
• A Survey of Exploration Methods in Reinforcement Learning
• Learning from Scarce Experience