Beyond Static Priors: Dynamic Neural Guidance for Large-Scale Ant Colony Optimization

2026-06-01

AI Summary

This work addresses the mismatch between static neural guidance and the dynamic, long-horizon nature of ant colony optimization (ACO) search, which leads to inconsistencies between training and inference. To resolve this, we propose DyNACO, the first framework that aligns neural guidance with ACO’s iterative process through dynamic adaptation: it periodically observes pheromone distributions and current best solutions to adjust guidance policies in real time. Coupled with a perturbation-based ACO backend and a range-constrained refinement mechanism, DyNACO ensures stable credit assignment and efficient solving at scale. On the Traveling Salesman Problem (TSP), it scales to 100,000 nodes, significantly outperforming existing neural methods while often reducing total runtime. On the Capacitated Vehicle Routing Problem (CVRP), it consistently surpasses unguided baselines with less than 1% neural overhead.

Abstract

Neural-guided Ant Colony Optimization (ACO) suffers from a fundamental training-inference misalignment: policies are typically trained to generate static priors (e.g., heatmaps), yet deployed to guide iterative, long-horizon search processes. In this paper, we present DyNACO, a novel framework that achieves dynamic neural guidance by periodically observing the pheromone distribution and the incumbent solution. To make DyNACO tractable at scale, we pair the policy with a perturbation-based ACO backend and a scope-restricted refinement mechanism that jointly ensure efficacy and stable credit assignment. On TSP, DyNACO scales to 100,000-node instances and outperforms neural baselines while often reducing total runtime compared to the unguided solver. We extend DyNACO to CVRP via a capacity-aware backend, consistently improving the unguided baseline with less than 1% neural overhead. We further provide in-depth analysis validating the model's generalization capabilities and elucidating why dynamic guidance outperforms static priors. Our work underscores the necessity of aligning neural training with iterative search dynamics in learning-guided optimization. The code is available at https://github.com/shoraaa/DyNACO.

Problem

• neural-guided optimization
• training-inference misalignment
• static priors
• iterative search
• dynamic guidance

Innovation

• Dynamic Neural Guidance
• Ant Colony Optimization
• Training-Inference Alignment
• Large-Scale Combinatorial Optimization
• Neural Heuristics