Threshold-aware Learning to Generate Feasible Solutions for Mixed Integer Programs

TL;DR

This paper introduces a threshold-aware learning method that optimizes variable assignment coverage in neural diving for mixed integer programs, significantly improving solution quality and speed over existing methods.

Contribution

It proposes a novel coverage optimization approach that aligns neural network predictions with MIP objectives, achieving state-of-the-art results in combinatorial optimization.

Findings

Achieves a 0.45% optimality gap in workload apportionment dataset.

Outperforms SCIP by ten times within one minute.

Demonstrates the effectiveness of coverage optimization in neural diving.

Abstract

Finding a high-quality feasible solution to a combinatorial optimization (CO) problem in a limited time is challenging due to its discrete nature. Recently, there has been an increasing number of machine learning (ML) methods for addressing CO problems. Neural diving (ND) is one of the learning-based approaches to generating partial discrete variable assignments in Mixed Integer Programs (MIP), a framework for modeling CO problems. However, a major drawback of ND is a large discrepancy between the ML and MIP objectives, i.e., variable value classification accuracy over primal bound. Our study investigates that a specific range of variable assignment rates (coverage) yields high-quality feasible solutions, where we suggest optimizing the coverage bridges the gap between the learning and MIP objectives. Consequently, we introduce a post-hoc method and a learning-based approach for optimizing the coverage. A key idea of our approach is to jointly learn to restrict the coverage search space and to predict the coverage in the learned search space. Experimental results demonstrate that learning a deep neural network to estimate the coverage for finding high-quality feasible solutions achieves state-of-the-art performance in NeurIPS ML4CO datasets. In particular, our method shows outstanding performance in the workload apportionment dataset, achieving the optimality gap of 0.45%, a ten-fold improvement over SCIP within the one-minute time limit.