Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Review
Explore this paper's citation graph
Summary
This article will discuss how a generalization of the reinforcement learning or optimal control problem, which is sometimes termed maximum entropy reinforcement learning, is equivalent to exact probabilistic inference in the case of deterministic dynamics, and variational inference inThe case of stochastic dynamics.
- Type
- preprint
- Published
- 2018-05-02
- Cited by
- 895
- References
- 64
- Access
- Open access
- OpenAlex
- https://openalex.org/W2799151646
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:19077536
Keywords
Reinforcement learning, Inference, Probabilistic logic, Reinforcement, Computer science
References
- Modeling Purposeful Adaptive Behavior with the Principle of Maximum Causal Entropy
- Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning
- Relative Entropy Policy Search
- On Stochastic Optimal Control and Reinforcement Learning by Approximate Inference (Extended Abstract)
- Inverse Optimal Control with Linearly-Solvable MDPs
- Covariant policy search
- Continuous Inverse Optimal Control with Locally Optimal Examples
- Maximum Entropy Deep Inverse Reinforcement Learning
- Expectation Propagation for approximate Bayesian inference
- Legibility and predictability of robot motion
- Function Optimization using Connectionist Reinforcement Learning Algorithms
- Goal-directed decision making as probabilistic inference: A computational framework and potential neural correlates
- Planning as inference.
- Variational Policy Search via Trajectory Optimization
- Maximum Entropy Inverse Reinforcement Learning
- Guided Policy Search
- Robot trajectory optimization using approximate inference
- Optimal control as a graphical model inference problem
- Reinforcement Learning: A Survey
- Reinforcement learning by reward-weighted regression for operational space control
Cited by
- Advances in Variational Inference
- Implicit Policy for Reinforcement Learning
- Variational Bayesian Reinforcement Learning with Regret Bounds
- Variational Inference with Tail-adaptive f-Divergence
- An Introduction to Probabilistic Programming
- Bayesian Transfer Reinforcement Learning with Prior Knowledge Rules
- Boosting Trust Region Policy Optimization by Normalizing Flows Policy
- Deep Imitative Models for Flexible Inference, Planning, and Control
- VIREL: A Variational Inference Framework for Reinforcement Learning
- Connecting the Dots Between MLE and RL for Sequence Generation
- On the Crossroad of Artificial Intelligence: A Revisit to Alan Turing and Norbert Wiener
- A new approach to learning in Dynamic Bayesian Networks (DBNs)
- Deconfounding Reinforcement Learning in Observational Settings
- Probabilistic Recursive Reasoning for Multi-Agent Reinforcement Learning
- Multi-Agent Generalized Recursive Reasoning
- Policy gradient methods on a multi-agent game with Kullback-Leibler costs
- End-to-End Robotic Reinforcement Learning without Reward Engineering
- A Multi-Hop Link Prediction Approach Based on Reinforcement Learning in Knowledge Graphs
- Optimal Control of Complex Systems through Variational Inference with a Discrete Event Decision Process
- TibGM: A Transferable and Information-Based Graphical Model Approach for Reinforcement Learning
Related papers
- Review on Reinforcement-concrete Bonded Anchorage
- Variable Ratio Reinforcement and Differential Reinforcement
- Reinforcement of inhibition.
- Choices between positive and negative reinforcement during treatment for escape-maintained behavior.
- REINFORCEMENT AND INSTRUCTIONS WITH MENTAL PATIENTS.
- Effects of rate of reinforcement-time upon concurrent operant performance.
- Effect of wall geometry on the behaviour of reinforced soil walls
- Inter-Related Effects of Experimental Set and External Reinforcement on Self-Reinforcement
- On evidence absorption for belief networks