Challenges and Opportunities in Offline Reinforcement Learning from Visual Observations.
Cong LuPhilip J. BallTim G. J. RudnerJack Parker-HolderMichael A. OsborneYee Whye TehPublished in: Trans. Mach. Learn. Res. (2023)
Keyphrases
- reinforcement learning
- lessons learned
- visual information
- visual features
- state space
- markov decision processes
- function approximation
- dynamic programming
- open issues
- real time
- low level
- high level
- case study
- learning algorithm
- real world
- visual search
- application scenarios
- robot control
- markov decision process
- key issues
- emerging technologies
- mobile robot
- optimal policy
- learning process
- multi agent systems
- video sequences
- multi agent
- web pages
- neural network
- data sets