A Distance Metric for Mixed Integer Programming Instances

TL;DR

This paper introduces the first mathematical distance metric for MILP instances, enabling better comparison and classification of problem instances, which improves solver guidance and instance set analysis.

Contribution

It proposes a novel, formulation-based distance metric for MILP instances, extending to instance-level comparison and outperforming existing methods in class identification.

Findings

The metric accurately identifies MILP instance classes.

The greedy variant is nearly as accurate as the exact version but much faster.

The method outperforms state-of-the-art non-learned approaches in class grouping tasks.

Abstract

Mixed-integer linear programming (MILP) is a powerful tool for addressing a wide range of real-world problems, but it lacks a clear structure for comparing instances. A reliable similarity metric could establish meaningful relationships between instances, enabling more effective evaluation of instance set heterogeneity and providing better guidance to solvers, particularly when machine learning is involved. Existing similarity metrics often lack precision in identifying instance classes or rely heavily on labeled data, which limits their applicability and generalization. To bridge this gap, this paper introduces the first mathematical distance metric for MILP instances, derived directly from their mathematical formulations. By discretizing right-hand sides, weights, and variables into classes, the proposed metric draws inspiration from the Earth mover's distance to quantify mismatches in weight-variable distributions for constraint comparisons. This approach naturally extends to enable instance-level comparisons. We evaluate both an exact and a greedy variant of our metric under various parameter settings, using the StrIPLIB dataset. Results show that all components of the metric contribute to class identification, and that the greedy version achieves accuracy nearly identical to the exact formulation while being nearly 200 times faster. Compared to state-of-the-art baselines, including feature-based, image-based, and neural network models, our unsupervised method consistently outperforms all non-learned approaches and rivals the performance of a supervised classifier on class and subclass grouping tasks.