SlateFree: a Model-Free Decomposition for Reinforcement Learning with Slate Actions.
Anastasios GiovanidisPublished in: CoRR (2022)
Keyphrases
- model free
- reinforcement learning
- reinforcement learning algorithms
- perceptual aliasing
- action selection
- temporal difference
- function approximation
- partially observable
- learning agent
- reward function
- state action
- action space
- rl algorithms
- optimal policy
- machine learning
- reinforcement learning methods
- state space
- policy iteration
- temporal difference learning
- policy evaluation
- learning algorithm
- markov decision processes
- average reward
- supervised learning
- pattern recognition
- impedance control
- function approximators
- training set
- e learning