Self-Improvement for Neural Combinatorial Optimization: Sample without Replacement, but Improvement

arXiv:2403.15180v1 Announce Type: new
Abstract: Current methods for end-to-end constructive neural combinatorial optimization usually train a policy using behavior cloning from expert solutions or policy gradient methods from reinforcement learning. While behavior cloning is straightforward, it requires expensive expert solutions, and policy gradient methods are often computationally demanding and complex to fine-tune. In this work, we bridge the two and simplify the training process by sampling multiple solutions for random instances using the current model in each epoch and then …

abstract arxiv behavior cloning cs.lg current expert gradient improvement optimization policy reinforcement reinforcement learning replacement sample self-improvement solutions train type