Towards An Unsupervised Learning Scheme for Efficiently Solving Parameterized Mixed-Integer Programs

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Qu, Shiyuan, Dong, Fenglian, Wei, Zhiwei, Shang, Chao
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915078007160832
author Qu, Shiyuan
Dong, Fenglian
Wei, Zhiwei
Shang, Chao
author_facet Qu, Shiyuan
Dong, Fenglian
Wei, Zhiwei
Shang, Chao
contents In this paper, we describe a novel unsupervised learning scheme for accelerating the solution of a family of mixed integer programming (MIP) problems. Distinct substantially from existing learning-to-optimize methods, our proposal seeks to train an autoencoder (AE) for binary variables in an unsupervised learning fashion, using data of optimal solutions to historical instances for a parametric family of MIPs. By a deliberate design of AE architecture and exploitation of its statistical implication, we present a simple and straightforward strategy to construct a class of cutting plane constraints from the decoder parameters of an offline-trained AE. These constraints reliably enclose the optimal binary solutions of new problem instances thanks to the representation strength of the AE. More importantly, their integration into the primal MIP problem leads to a tightened MIP with the reduced feasible region, which can be resolved at decision time using off-the-shelf solvers with much higher efficiency. Our method is applied to a benchmark batch process scheduling problem formulated as a mixed integer linear programming (MILP) problem. Comprehensive results demonstrate that our approach significantly reduces the computational cost of off-the-shelf MILP solvers while retaining a high solution quality. The codes of this work are open-sourced at https://github.com/qushiyuan/AE4BV.
format Preprint
id arxiv_https___arxiv_org_abs_2412_17623
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Towards An Unsupervised Learning Scheme for Efficiently Solving Parameterized Mixed-Integer Programs
Qu, Shiyuan
Dong, Fenglian
Wei, Zhiwei
Shang, Chao
Optimization and Control
Machine Learning
In this paper, we describe a novel unsupervised learning scheme for accelerating the solution of a family of mixed integer programming (MIP) problems. Distinct substantially from existing learning-to-optimize methods, our proposal seeks to train an autoencoder (AE) for binary variables in an unsupervised learning fashion, using data of optimal solutions to historical instances for a parametric family of MIPs. By a deliberate design of AE architecture and exploitation of its statistical implication, we present a simple and straightforward strategy to construct a class of cutting plane constraints from the decoder parameters of an offline-trained AE. These constraints reliably enclose the optimal binary solutions of new problem instances thanks to the representation strength of the AE. More importantly, their integration into the primal MIP problem leads to a tightened MIP with the reduced feasible region, which can be resolved at decision time using off-the-shelf solvers with much higher efficiency. Our method is applied to a benchmark batch process scheduling problem formulated as a mixed integer linear programming (MILP) problem. Comprehensive results demonstrate that our approach significantly reduces the computational cost of off-the-shelf MILP solvers while retaining a high solution quality. The codes of this work are open-sourced at https://github.com/qushiyuan/AE4BV.
title Towards An Unsupervised Learning Scheme for Efficiently Solving Parameterized Mixed-Integer Programs
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2412.17623