We present the Generative Flow Ant Colony Sampler (GFACS), a novel meta-heuristic method that hierarchically combines amortized inference and parallel stochastic search. Our method first leverages Generative Flow Networks (GFlowNets) to amortize a \emph{multi-modal} prior distribution over combinatorial solution space that encompasses both high-reward and diversified solutions. This prior is iteratively updated via parallel stochastic search in the spirit of Ant Colony Optimization (ACO), leading to the posterior distribution that generates near-optimal solutions. Extensive experiments across seven combinatorial optimization problems demonstrate GFACS's promising performances.
Cite
@article{arxiv.2403.07041,
title = {Ant Colony Sampling with GFlowNets for Combinatorial Optimization},
author = {Minsu Kim and Sanghyeok Choi and Hyeonah Kim and Jiwoo Son and Jinkyoo Park and Yoshua Bengio},
journal= {arXiv preprint arXiv:2403.07041},
year = {2025}
}