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

Mathematical Programming Computation

QPLIB: a library of quadratic programming instances

Fabio Furini, Emiliano Traversi, Pietro Belotti, Antonio Frangioni, Ambros Gleixner, Nick Gould, Leo Liberti, Andrea Lodi, Ruth Misener, Hans Detlef Mittelmann, Nikolaos V. Sahinidis, Stefan Vigerske, Angelika Wiegele

Published Sep 22, 2018• 104 citations
Publisher
DOI
10.1007/s12532-018-0147-4

Related papers

Local Search for Integer Quadratic Programming

Xiang He, Peng Lin, Tao Jiang +1 more

Integer quadratic programming (IQP) is an important problem in operations research. Local search is a powerful primal heuristic method for solving hard problems, but the research on local search algorithms for solving IQP is still in its early stages. This paper introduces an efficient local search algorithm called LS-IQCQP, which stands for “local search for integer quadratically constrained quadratic programming.” The algorithm is designed for solving general IQP, and it is capable of handling cases where quadratic terms appear in the objective function, constraints, or both. We propose four newly designed operators: the root operator for feasibility and three operators—inequality exploration, equality balance, and unconstrained free operator—for optimizing the objective function. To evaluate the quality of these operations, a weighted scoring function is introduced. Based on the scoring function, LS-IQCQP iteratively applies effective operations and employs a WalkSAT-like approach (a classic local search algorithm for satisfiability problems) by randomly selecting an operation to escape local optima. Experiments are conducted on standard IQP benchmarks from the Quadratic Programming Library (QPLIB) and Mixed-Integer Nonlinear Programming Library (MINLPLIB), comparing LS-IQCQP with several state-of-the-art IQP solvers. Experimental results demonstrate that LS-IQCQP is competitive with the most powerful commercial solver Gurobi and other state-of-the-art solvers in terms of finding a good feasible solution quickly. Moreover, LS-IQCQP has established six new records for QPLIB and MINLPLIB open instances by finding the new best-known solutions. History: Accepted by Erwin Pesch, Area Editor for Heuristic Search & Approximation Algorithms. Funding: This research was supported by the National Key Research and Development Program of China [Grant 2023YFA1009500]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2025.1178 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2025.1178 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .

arXiv:2409.19668•INFORMS journal on computing•May 12, 2026•3 citations

Approximation Algorithms for Quadratic Programming

Minyue Fu, Zhi-Quan Tom Luo, Yinyu Ye

Journal of Combinatorial Optimization•Mar 1, 1998•78 citations

An Alternating Primal Heuristic for Nonconvex MIQCQP with Dynamic Convexification and Parallel Local Branching

Yongzheng Dai, Chen Chen

We develop a novel primal heuristic for nonconvex Mixed-Integer Quadratically Constrained Quadratic Programs (MIQCQPs). The method is built around a convex approximation that is dynamically adjusted within a feasibility-pump-style alternating heuristic. Approximations are adjusted based on the structure of the MIQCQP instance. Additionally, parallelized local branching is incorporated to further refine detected solutions. This paper builds upon the second-place finalist submission in the 2025 Land-Doig MIP Computational Competition. Our results are validated with computational experiments on instances from QPLIB, finding feasible solutions for three previously unsolved cases and improving the best-known solutions for fifteen instances within five minutes of runtime.

arXiv:2604.04417•Apr 6, 2026•0 citations

Approximating quadratic programming with bound and quadratic constraints

Yinyu Ye

Mathematical Programming•Feb 1, 1999•170 citations

Quadratic Unconstrained Binary Optimization (QUBO) Modeling

WORLD SCIENTIFIC eBooks•Mar 20, 2023•3 citations

Learning to Select Between LP and SDP Relaxations for Nonconvex Quadratically Constrained Quadratic Programs

Dedeoglu, Muge, Ozen, Buket, Burak Kocuk

Quadratically constrained quadratic programs (QCQPs) are ubiquitous in optimization: Such problems arise in applications from operations research, power systems, signal processing, chemical engineering, and portfolio theory, among others. Despite their flexibility in modeling real-life situations and the recent effort to understand their properties, nonconvex QCQPs are hard to solve in practice. Most of the approaches in the literature are based on either Linear Programming (LP) or Semidefinite Programming (SDP) relaxations, each of which works very well for some problem subclasses but perform poorly on others. In this paper, we develop a relaxation selection procedure for nonconvex QCQPs that can adaptively decide whether an LP- or SDP-based approach is expected to be more beneficial by considering the instance structure. The proposed methodology relies on utilizing machine learning methods that involve features derived from spectral properties and sparsity patterns of data matrices, and once trained appropriately within a dimension-independent setup, the prediction model applies to any instance with an arbitrary number of variables and constraints. We develop classification and regression models under different feature-design setups, including a dimension-independent representation, and evaluate them on synthetically generated instances and a combined benchmark dataset consisting of MINLPLib and QPLIB instances. We further investigate relaxation selection within a spatial branch-and-bound framework, evaluating the trade-off between lower bound quality and computational time.

arXiv:2501.03954•Jan 7, 2025•1 citation
Integer quadratic programming (IQP) is an important problem in operations research. Local search is a powerful primal heuristic method for solving hard problems, but the research on local search algorithms for solving IQP is still in its early stages. This paper introduces an efficient local search algorithm called LS-IQCQP, which stands for “local search for integer quadratically constrained quadratic programming.” The algorithm is designed for solving general IQP, and it is capable of handling cases where quadratic terms appear in the objective function, constraints, or both. We propose four newly designed operators: the root operator for feasibility and three operators—inequality exploration, equality balance, and unconstrained free operator—for optimizing the objective function. To evaluate the quality of these operations, a weighted scoring function is introduced. Based on the scoring function, LS-IQCQP iteratively applies effective operations and employs a WalkSAT-like approach (a classic local search algorithm for satisfiability problems) by randomly selecting an operation to escape local optima. Experiments are conducted on standard IQP benchmarks from the Quadratic Programming Library (QPLIB) and Mixed-Integer Nonlinear Programming Library (MINLPLIB), comparing LS-IQCQP with several state-of-the-art IQP solvers. Experimental results demonstrate that LS-IQCQP is competitive with the most powerful commercial solver Gurobi and other state-of-the-art solvers in terms of finding a good feasible solution quickly. Moreover, LS-IQCQP has established six new records for QPLIB and MINLPLIB open instances by finding the new best-known solutions. History: Accepted by Erwin Pesch, Area Editor for Heuristic Search & Approximation Algorithms. Funding: This research was supported by the National Key Research and Development Program of China [Grant 2023YFA1009500]. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2025.1178 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2025.1178 ). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/ .
We develop a novel primal heuristic for nonconvex Mixed-Integer Quadratically Constrained Quadratic Programs (MIQCQPs). The method is built around a convex approximation that is dynamically adjusted within a feasibility-pump-style alternating heuristic. Approximations are adjusted based on the structure of the MIQCQP instance. Additionally, parallelized local branching is incorporated to further refine detected solutions. This paper builds upon the second-place finalist submission in the 2025 Land-Doig MIP Computational Competition. Our results are validated with computational experiments on instances from QPLIB, finding feasible solutions for three previously unsolved cases and improving the best-known solutions for fifteen instances within five minutes of runtime.
Quadratically constrained quadratic programs (QCQPs) are ubiquitous in optimization: Such problems arise in applications from operations research, power systems, signal processing, chemical engineering, and portfolio theory, among others. Despite their flexibility in modeling real-life situations and the recent effort to understand their properties, nonconvex QCQPs are hard to solve in practice. Most of the approaches in the literature are based on either Linear Programming (LP) or Semidefinite Programming (SDP) relaxations, each of which works very well for some problem subclasses but perform poorly on others. In this paper, we develop a relaxation selection procedure for nonconvex QCQPs that can adaptively decide whether an LP- or SDP-based approach is expected to be more beneficial by considering the instance structure. The proposed methodology relies on utilizing machine learning methods that involve features derived from spectral properties and sparsity patterns of data matrices, and once trained appropriately within a dimension-independent setup, the prediction model applies to any instance with an arbitrary number of variables and constraints. We develop classification and regression models under different feature-design setups, including a dimension-independent representation, and evaluate them on synthetically generated instances and a combined benchmark dataset consisting of MINLPLib and QPLIB instances. We further investigate relaxation selection within a spatial branch-and-bound framework, evaluating the trade-off between lower bound quality and computational time.