On the benefits of pixel-based hierarchical policies for task generalization.
Tudor Cristea-PlatonBogdan MazoureJosh SusskindWalter TalbottPublished in: CoRR (2024)
Keyphrases
- hierarchical reinforcement learning
- management system
- unsupervised learning
- optimal policy
- neural network
- information retrieval
- knowledge base
- reinforcement learning
- multi agent systems
- artificial neural networks
- dynamic programming
- hierarchical clustering
- hierarchical model
- hierarchical classification
- hierarchical representation
- control policies