The Value of Reward Lookahead in Reinforcement Learning.
Nadav MerlisDorian BaudryVianney PerchetPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- state space
- reinforcement learning algorithms
- eligibility traces
- reward function
- machine learning
- learning algorithm
- action selection
- model free
- optimal policy
- reward shaping
- multi agent
- average reward
- optimal control
- partially observable
- learning agent
- reinforcement learning methods
- policy search
- robotic control
- total reward
- partially observable environments
- control policy
- temporal difference
- learning problems
- transfer learning
- supervised learning