Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2603.19318 · Lecture notes in computer science

Towards Solving Polynomial-Objective Integer Programming with Hypergraph Neural Networks

Minshuo Li, Yizhi Wu, Pavel Troubil, Yingqian Zhang, Wim Nuijten

Published Jan 1, 2026• 0 citations
arXivHTMLPDF
DOI
10.1007/978-3-032-27242-3_18

Related papers

Reusability report: A distributed strategy for solving combinatorial optimization problems with hypergraph neural networks

Xiaodi Li, Jianfeng Gui, Wei Xue +5 more

Nature Machine Intelligence•Nov 19, 2025•4 citations

A polynomial case of convex integer quadratic programming problems with box integer constraints

Chunli Liu, Jianjun Gao

Journal of Global Optimization•Jan 6, 2015•5 citations

Globally solving nonconvex quadratic programming problems with box constraints via integer programming methods

Pierre Bonami, Oktay Günlük, Jeff Linderoth

Mathematical Programming Computation•Feb 24, 2018•40 citations

A Polynomial Time Solvable Algorithm to Binary Quadratic Programming Problems with Q Being a Seven-Diagonal Matrix and Its Neural Network Implementation

Shenshen Gu, Peng Jiao, Rui Cui

Lecture notes in computer science•Jan 1, 2014•2 citations

Combinatorial optimization with physics-inspired graph neural networks

Martin J. A. Schuetz, John Kyle Brubaker, Helmut G. Katzgraber

arXiv:2107.01188•Nature Machine Intelligence•Apr 21, 2022•236 citations

BIPNN: Learning to Solve Binary Integer Programming via Hypergraph Neural Networks

Sen Bai, Chunqi Yang, Xin Wei Bai +2 more

Binary (0-1) integer programming (BIP) is pivotal in scientific domains requiring discrete decision-making. As the advance of AI computing, recent works explore neural network-based solvers for integer linear programming (ILP) problems. Yet, they lack scalability for tackling nonlinear challenges. To handle nonlinearities, state-of-the-art Branch-and-Cut solvers employ linear relaxations, leading to exponential growth in auxiliary variables and severe computation limitations. To overcome these limitations, we propose BIPNN (Binary Integer Programming Neural Network), an unsupervised learning framework to solve nonlinear BIP problems via hypergraph neural networks (HyperGNN). Specifically, BIPNN reformulates BIPs-constrained, discrete, and nonlinear (sin, log, exp) optimization problems-into unconstrained, differentiable, and polynomial loss functions. The reformulation stems from the observation of a precise one-to-one mapping between polynomial BIP objectives and hypergraph structures, enabling the unsupervised training of HyperGNN to optimize BIP problems in an end-to-end manner. On this basis, we propose a GPU-accelerated and continuous-annealing-enhanced training pipeline for BIPNN. The pipeline enables BIPNN to optimize large-scale nonlinear terms in BIPs fully in parallel via straightforward gradient descent, thus significantly reducing the training cost while ensuring the generation of discrete, high-quality solutions. Extensive experiments on synthetic and real-world datasets highlight the superiority of our approach.

arXiv:2505.20997•neural information processing systems•Jan 1, 2025•1 citation
Binary (0-1) integer programming (BIP) is pivotal in scientific domains requiring discrete decision-making. As the advance of AI computing, recent works explore neural network-based solvers for integer linear programming (ILP) problems. Yet, they lack scalability for tackling nonlinear challenges. To handle nonlinearities, state-of-the-art Branch-and-Cut solvers employ linear relaxations, leading to exponential growth in auxiliary variables and severe computation limitations. To overcome these limitations, we propose BIPNN (Binary Integer Programming Neural Network), an unsupervised learning framework to solve nonlinear BIP problems via hypergraph neural networks (HyperGNN). Specifically, BIPNN reformulates BIPs-constrained, discrete, and nonlinear (sin, log, exp) optimization problems-into unconstrained, differentiable, and polynomial loss functions. The reformulation stems from the observation of a precise one-to-one mapping between polynomial BIP objectives and hypergraph structures, enabling the unsupervised training of HyperGNN to optimize BIP problems in an end-to-end manner. On this basis, we propose a GPU-accelerated and continuous-annealing-enhanced training pipeline for BIPNN. The pipeline enables BIPNN to optimize large-scale nonlinear terms in BIPs fully in parallel via straightforward gradient descent, thus significantly reducing the training cost while ensuring the generation of discrete, high-quality solutions. Extensive experiments on synthetic and real-world datasets highlight the superiority of our approach.