Feature Augmentation of GNNs for ILPs: Local Uniqueness Suffices
About
Integer Linear Programs (ILPs) are central to real-world optimizations but notoriously difficult to solve. Learning to Optimize (L2O) has emerged as a promising paradigm, with Graph Neural Networks (GNNs) serving as the standard backbone. However, standard anonymous GNNs are limited in expressiveness for ILPs, and the common enhancement of augmenting nodes with globally unique identifiers (UIDs) typically introduces spurious correlations that severely harm generalization. To address this tradeoff, we propose a parsimonious Local-UID scheme based on d-hop uniqueness coloring, which ensures identifiers are unique only within each node's d-hop neighborhood. Building on this scheme, we introduce ColorGNN, which incorporates color information via color-conditioned embeddings, and ColorUID, a lightweight feature-level variant. We prove that for d-layer networks, Local-UIDs achieve the expressive power of Global-UIDs while offering stronger generalization. Extensive experiments show that our approach yields substantial and robust gains across ILP benchmarks.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Integer Linear Programming Solving | CA | -- | 8 | |
| Integer Linear Programming Solving | SMSP | Optimality Gap 30%0.23 | 7 | |
| Integer Linear Programming Solving | BPP | Metric at 30% Gap0.00e+0 | 7 | |
| Integer Linear Programming Solving | BIP | Performance at 30th Percentile2.73 | 7 | |
| Integer Linear Programming Solving | IS | Performance (30%)0.00e+0 | 5 |