Provably Efficient Learning in Partially Observable Contextual Bandit.
Xueping GongJiheng ZhangPublished in: CoRR (2023)
Keyphrases
- efficient learning
- partially observable
- contextual bandit
- upper confidence bound
- state space
- decision problems
- dynamical systems
- reinforcement learning
- markov decision processes
- partial observations
- infinite horizon
- news recommendation
- partial observability
- belief state
- planning domains
- learning algorithm
- pattern languages
- membership queries
- reward function
- data mining