Deep Reinforcement Learning with Dual Targeting Algorithm.
Naoki KodamaTaku HaradaKazuteru MiyazakiPublished in: IJCNN (2019)
Keyphrases
- reinforcement learning
- dynamic programming
- detection algorithm
- optimization algorithm
- preprocessing
- learning algorithm
- times faster
- improved algorithm
- k means
- theoretical analysis
- computationally efficient
- objective function
- optimal solution
- clustering method
- neural network
- model free
- expectation maximization
- particle swarm optimization
- worst case
- experimental evaluation
- cost function
- computational complexity
- linear programming
- least squares
- optimal policy
- tree structure
- matching algorithm
- convergence rate
- recognition algorithm
- data sets
- gradient projection