Generalized Maximum Entropy Reinforcement Learning via Reward Shaping.
Feng TaoMingkang WuYongcan CaoPublished in: IEEE Trans. Artif. Intell. (2024)
Keyphrases
- maximum entropy
- reward shaping
- reinforcement learning
- reinforcement learning algorithms
- complex domains
- maximum entropy principle
- markov models
- markov decision problems
- random fields
- function approximation
- state space
- principle of maximum entropy
- conditional random fields
- model free
- multi agent
- transformation based learning
- learning algorithm
- minimum cross entropy
- reward function
- temporal difference
- markov decision processes
- dynamic programming
- machine learning
- optimal control
- pairwise
- bayesian networks