Revisiting Peng's Q(λ) for Modern Reinforcement Learning.
Tadashi KozunoYunhao TangMark RowlandRémi MunosSteven KapturowskiWill DabneyMichal ValkoDavid AbelPublished in: ICML (2021)
Keyphrases
- reinforcement learning
- function approximation
- state space
- reinforcement learning algorithms
- multi agent
- optimal policy
- robotic control
- temporal difference
- real world
- action selection
- model free
- databases
- transition model
- robot control
- optimal control
- learning process
- decision making
- neural network
- sufficient conditions
- evolutionary algorithm
- natural language
- learning algorithm
- genetic algorithm
- learning capabilities
- machine learning
- temporal difference learning
- evolutionary learning
- real time