Latent Guided Sampling for Combinatorial Optimization
This work addresses computational bottlenecks in combinatorial optimization for applications such as logistics and manufacturing, offering an incremental improvement over existing neural methods.
The paper tackled the challenge of combinatorial optimization in domains like logistics and drug discovery by proposing LGS-Net, a latent space model with Latent Guided Sampling, which achieved state-of-the-art performance among RL-based approaches on benchmark routing tasks.
Combinatorial Optimization problems are widespread in domains such as logistics, manufacturing, and drug discovery, yet their NP-hard nature makes them computationally challenging. Recent Neural Combinatorial Optimization methods leverage deep learning to learn solution strategies, trained via Supervised or Reinforcement Learning (RL). While promising, these approaches often rely on task-specific augmentations, perform poorly on out-of-distribution instances, and lack robust inference mechanisms. Moreover, existing latent space models either require labeled data or rely on pre-trained policies. In this work, we propose LGS-Net, a novel latent space model that conditions on problem instances, and introduce an efficient inference method, Latent Guided Sampling (LGS), based on Markov Chain Monte Carlo and Stochastic Approximation. We show that the iterations of our method form a time-inhomogeneous Markov Chain and provide rigorous theoretical convergence guarantees. Empirical results on benchmark routing tasks show that our method achieves state-of-the-art performance among RL-based approaches.