Relate to Predict: Towards Task-Independent Knowledge Representations for Reinforcement Learning.
Thomas SchnürerMalte ProbstHorst-Michael GrossPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- function approximation
- state space
- robot control
- search engine
- reinforcement learning algorithms
- supervised learning
- optimal policy
- function approximators
- real time
- control problems
- optimal control
- markov decision processes
- monte carlo
- multi agent
- e learning
- social networks
- genetic algorithm
- information retrieval
- databases