Solving Sokoban with forward-backward reinforcement learning.
Yaron ShohamGal ElidanPublished in: CoRR (2021)
Keyphrases
- forward backward
- reinforcement learning
- hidden markov models
- function approximation
- markov decision problems
- combinatorial optimization
- dynamic programming
- supervised learning
- optimal policy
- transfer learning
- problems in artificial intelligence
- neural network
- robotic control
- transition model
- markov decision process
- learning capabilities
- state space
- multi agent
- website