Discovering and Removing Exogenous State Variables and Rewards for Reinforcement Learning.
Thomas G. DietterichGeorge TrimponiasZhitang ChenPublished in: CoRR (2018)
Keyphrases
- state variables
- reinforcement learning
- state space
- reward function
- markov decision processes
- dynamic systems
- reinforcement learning algorithms
- function approximation
- optimal policy
- heuristic search
- dynamic bayesian networks
- model free
- multi agent
- dynamic programming
- machine learning
- markov decision process
- autoregressive
- partially observable
- random variables
- particle filter
- reward shaping
- policy iteration
- temporal difference
- adaptive control
- optimal control
- planning problems
- learning algorithm
- action selection
- dynamical systems
- action space
- function approximators
- markov decision problems
- data mining