Pre-training of Deep RL Agents for Improved Learning under Domain Randomization.
Artemij AmiranashviliMax ArgusLukás HermannWolfram BurgardThomas BroxPublished in: CoRR (2021)
Keyphrases
- learned knowledge
- learning process
- reinforcement learning
- multi agent
- learning agents
- active learning
- learning algorithm
- online learning
- supervised learning
- complex domains
- action selection
- multi agent systems
- knowledge acquisition
- multiagent systems
- feedforward neural networks
- deep architectures
- computer based training
- decision theoretic
- domain specific
- cooperative
- dynamic environments
- multiagent learning
- decision making