Bootstrapped fitness critics with bidirectional temporal difference.
Golden RockefellerKagan TumerPublished in: GECCO Companion (2022)
Keyphrases
- temporal difference
- td learning
- reinforcement learning
- function approximation
- evaluation function
- monte carlo
- genetic programming
- temporal difference learning
- model free
- genetic algorithm
- fitness function
- reinforcement learning algorithms
- evolutionary algorithm
- step size
- action selection
- temporal difference methods
- policy iteration
- multiscale
- data mining
- particle swarm optimization
- supervised learning
- decision trees
- function approximators
- neural network
- data sets