Explore-Go: Leveraging Exploration for Generalisation in Deep Reinforcement Learning.
Max WeltevredeFelix KaubekMatthijs T. J. SpaanWendelin BöhmerPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- exploration strategy
- action selection
- active exploration
- function approximation
- model based reinforcement learning
- autonomous learning
- exploration exploitation
- learning algorithm
- multi agent
- state space
- optimal policy
- model free
- policy search
- multi agent reinforcement learning
- learning process
- real world
- temporal difference
- reward function
- reinforcement learning algorithms
- markov decision process
- deep learning
- temporal difference learning
- learning problems
- dynamic programming
- evolutionary algorithm
- multi agent systems
- genetic algorithm
- machine learning