Improving Generalization of Deep Reinforcement Learning-based TSP Solvers.
Wenbin OuyangYisen WangShaochen HanZhejian JinPaul WengPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- traveling salesman problem
- function approximation
- markov decision processes
- learning algorithm
- model free
- dynamic programming
- state space
- supervised learning
- search space
- action selection
- travelling salesman
- transfer learning
- learning process
- combinatorial optimization
- learning problems
- multi agent
- data sets
- temporal difference
- deep learning
- sat solving