Value-based Algorithms Optimization with Discounted Multiple-step Learning Method in Deep Reinforcement Learning.
Haibo DengShiqun YinXiaohong DengShiwei LiPublished in: HPCC/DSS/SmartCity (2020)
Keyphrases
- reinforcement learning
- learning algorithm
- combinatorial optimization
- optimization algorithm
- computationally efficient
- theoretical analysis
- dynamic programming
- model free
- significant improvement
- cost function
- prior knowledge
- optimization method
- learning process
- computational cost
- optimization problems
- markov decision processes
- fitted q iteration
- optimization process
- neural network
- supervised learning
- multi objective
- computational complexity
- machine learning algorithms
- unsupervised learning
- objective function
- learning problems
- optimization methods
- constrained optimization
- optimization procedure
- support vector machine
- function approximators
- reinforcement learning methods
- automatically learned