A First-Occupancy Representation for Reinforcement Learning.
Ted MoskovitzSpencer R. WilsonManeesh SahaniPublished in: ICLR (2022)
Keyphrases
- reinforcement learning
- information systems
- neural network
- optimal policy
- computer vision
- dynamic programming
- case study
- visual representation
- feature representation
- supervised learning
- state space
- probabilistic model
- data structure
- real time
- e learning
- feature selection
- artificial intelligence
- learning algorithm
- databases
- data sets