Discovering Diverse Solutions in Deep Reinforcement Learning.
Takayuki OsaVoot TangkarattMasashi SugiyamaPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- function approximation
- model free
- optimal solution
- state space
- learning algorithm
- machine learning
- similar problems
- benchmark problems
- learning problems
- wide variety
- least squares
- linear programming
- optimal policy
- dynamic programming
- feasible solution
- objective function
- solution quality
- case study
- information systems
- control strategies
- action selection
- temporal difference
- real world