Slowly Changing Adversarial Bandit Algorithms are Efficient for Discounted MDPs.
Ian A. KashLev ReyzinZishun YuPublished in: ALT (2024)
Keyphrases
- markov decision processes
- computationally efficient
- significant improvement
- computationally intensive
- computationally expensive
- optimization problems
- bandit problems
- computationally complex
- parallel architectures
- pruning strategy
- least squares
- computational cost
- computational complexity
- reinforcement learning
- learning algorithm
- state space
- theoretical analysis
- association rules
- data structure
- data mining