Efficient Bayes-Adaptive Reinforcement Learning using Sample-Based Search

Abstract

Bayesian model-based reinforcement learning is a formally elegant approach to learning optimal behaviour under model uncertainty, trading off exploration and exploitation in an ideal way. Unfortunately, finding the resulting Bayes-optimal policies is notoriously taxing, since the search space becomes enormous. In this paper we introduce a tractable, sample-based method for approximate Bayes-optimal planning which exploits Monte-Carlo tree search. Our approach outperformed prior Bayesian model-based RL algorithms by a significant margin on several well-known benchmark problems -- because it avoids expensive applications of Bayes rule within the search tree by lazily sampling models from the current beliefs. We illustrate the advantages of our approach by showing it working in an infinite state space domain which is qualitatively out of reach of almost all previous work in Bayesian exploration.

14 pages, 7 figures, includes supplementary material. Advances in Neural Information Processing Systems (NIPS) 2012

References in corpus (3)

• Bandit Algorithms for Tree Search
• A Bayesian Sampling Approach to Exploration in Reinforcement Learning
• Learning is planning: near Bayes-optimal reinforcement learning via Monte-Carlo tree search

Cited by in corpus (32)

• Count-Based Exploration with Neural Density Models
• Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models
• (More) Efficient Reinforcement Learning via Posterior Sampling
• VariBAD: A Very Good Method for Bayes-Adaptive Deep RL via Meta-Learning
• The Uncertainty Bellman Equation and Exploration
• Hidden Parameter Markov Decision Processes: A Semiparametric Regression Approach for Discovering Latent Task Parametrizations
• AutoCompress: An Automatic DNN Structured Pruning Framework for Ultra-High Compression Rates
• Exploratory Gradient Boosting for Reinforcement Learning in Complex Domains
• Efficient exploration with Double Uncertain Value Networks
• Making Sense of Reinforcement Learning and Probabilistic Inference
• Benchmarking for Bayesian Reinforcement Learning
• Offline Meta Learning of Exploration
• Assumed Density Filtering Q-learning
• Blazing the trails before beating the path: Sample-efficient Monte-Carlo planning
• The Potential of the Return Distribution for Exploration in RL
• Risk-Averse Bayes-Adaptive Reinforcement Learning
• A Survey of Exploration Methods in Reinforcement Learning
• Bayesian Policy Optimization for Model Uncertainty
• Bayesian Reinforcement Learning via Deep, Sparse Sampling
• Scalable Planning and Learning for Multiagent POMDPs: Extended Version
• Hyper-parameter Tuning under a Budget Constraint
• Robust Stochastic Bayesian Games for Behavior Space Coverage
• Learning in POMDPs with Monte Carlo Tree Search
• Fast Exploration with Simplified Models and Approximately Optimistic Planning in Model Based Reinforcement Learning
• Better Optimism By Bayes: Adaptive Planning with Rich Models
• Posterior Sampling for Anytime Motion Planning on Graphs with Expensive-to-Evaluate Edges
• TTR-Based Reward for Reinforcement Learning with Implicit Model Priors
• The Assistive Multi-Armed Bandit
• Knowledge is reward: Learning optimal exploration by predictive reward cashing
• Bayes-CPACE: PAC Optimal Exploration in Continuous Space Bayes-Adaptive Markov Decision Processes
• Deep Reinforced Attention Regression for Partial Sketch Based Image Retrieval
• Parameterized Indexed Value Function for Efficient Exploration in Reinforcement Learning