Extension of constraint-procedural logic-generated environments for deep Q-learning agent training and benchmarking.
Giovanni De GasperisStefania CostantiniAndrea RafanelliPatrizio MigliariniIvan LetteriAbeer DyoubPublished in: J. Log. Comput. (2023)
Keyphrases
- learning agent
- subject matter experts
- reinforcement learning
- dynamic environments
- state space
- selective perception
- path planning
- supervised learning
- solving problems
- reinforcement learning algorithms
- learning capabilities
- training set
- knowledge representation
- random walk
- dynamic programming
- learning strategies
- single agent
- learning algorithm
- neural network