Unsupervised State Representation Learning in Partially Observable Atari Games.
Li MengMorten GoodwinAnis YazidiPaal EngelstadPublished in: CAIP (2) (2023)
Keyphrases
- partially observable
- reinforcement learning
- state space
- markov decision processes
- learning algorithm
- action models
- dynamical systems
- partial observations
- decision problems
- stochastic games
- heuristic search
- supervised learning
- infinite horizon
- semi supervised
- optimal solution
- markov decision problems
- decision making