The Ladder in Chaos: A Simple and Effective Improvement to General DRL Algorithms by Policy Path Trimming and Boosting.
Hongyao TangMin ZhangJianye HaoPublished in: CoRR (2023)
Keyphrases
- learning algorithm
- special case
- computationally efficient
- black box
- orders of magnitude
- significant improvement
- optimization problems
- inverse reinforcement learning
- straight forward
- computational complexity
- computational cost
- optimal policy
- machine learning
- computationally tractable
- neural network
- computationally hard
- data sets
- data structure
- ensemble classifier
- policy iteration
- decision tree algorithm