Towards Understanding Distributional Reinforcement Learning: Regularization, Optimization, Acceleration and Sinkhorn Algorithm.
Ke SunYingnan ZhaoYi LiuEnze ShiYafei WangAref SadeghiXiaodong YanBei JiangLinglong KongPublished in: CoRR (2021)
Keyphrases
- optimization algorithm
- learning algorithm
- optimization method
- detection algorithm
- stochastic approximation
- reinforcement learning
- computational complexity
- simulated annealing
- cost function
- machine learning
- computational cost
- dynamic programming
- np hard
- k means
- optimal solution
- multi objective
- matching algorithm
- recognition algorithm
- optimization process
- particle swarm optimization
- expectation maximization
- monte carlo
- objective function
- regularization term
- genetic algorithm