Instruction for reinforcement learning agent based on sub-rewards and forgetting.
Toshihiko WatanabeToru SawaPublished in: FUZZ-IEEE (2010)
Keyphrases
- reinforcement learning
- multi agent
- learning capabilities
- markov decision processes
- function approximation
- state space
- reinforcement learning algorithms
- machine learning
- reward function
- instructional design
- learning algorithm
- multi agent systems
- policy iteration
- model free
- reward shaping
- optimal policy
- neural network
- temporal difference
- learning process
- computer technology
- incremental learning
- transfer learning
- supervised learning
- learning agents
- online learning
- total reward
- instruction set
- robotic control
- computer software
- multimedia
- action selection
- cooperative learning
- optimal control