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

IEICE Technical Report; IEICE Tech. Rep.

QUBO solver for combinatorial optimization problems with constraints

Fumiyo Takano, Motoi Suzuki, Y. Kobayashi, Takuya Araki

Published Nov 21, 2019• 7 citations
OpenAlex

Related papers

QUBO Model Formulation Based on Petri Net Behavioral Description for Combinatorial Optimization Problems

Keisuke Tokuhira, Morikazu Nakamura, Mitsunaga Kinjo +1 more

Lecture notes in networks and systems•Oct 20, 2022•0 citations

HyCiM: A Hybrid Computing-in-Memory QUBO Solver for General Combinatorial Optimization Problems with Inequality Constraints

Yu Qian, Zeyu Yang, Kai Ni +3 more

Computationally challenging combinatorial optimization problems (COPs) play a fundamental role in various applications. To tackle COPs, many Ising machines and Quadratic Unconstrained Binary Optimization (QUBO) solvers have been proposed, which typically involve direct transformation of COPs into Ising models or equivalent QUBO forms (D-QUBO). However, when addressing COPs with inequality constraints, this D-QUBO approach introduces numerous extra auxiliary variables, resulting in a substantially larger search space, increased hardware costs, and reduced solving efficiency. In this work, we propose HyCiM, a novel hybrid computing-inmemory (CiM) based QUBO solver framework, designed to overcome aforementioned challenges. The proposed framework consists of (i) an innovative transformation method (first to our known) that converts COPs with inequality constraints into an inequality-QUBO form, thus eliminating the need of expensive auxiliary variables and associated calculations; (ii) "inequality filter", a ferroelectric FET (FeFET)-based CiM circuit that accelerates the inequality evaluation, and filters out infeasible input configurations; (iii) a FeFET-based CiM annealer that is capable of approaching global solutions of COPs via iterative QUBO computations within a simulated annealing process. The evaluation results show that HyCiM drastically narrows down the search space, eliminating 2100 to 22536 infeasible input configurations compared to the conventional D-QUBO approach. Consequently, the narrowed search space, reduced to 2100 feasible input configurations, leads to a substantial hardware area overhead reduction, ranging from 88.06% to 99.96%. Additionally, HyCiM consistently exhibits a high solving efficiency, achieving a remarkable average success rate of 98.54%, whereas D-QUBO implementatoin shows only 10.75%.

arXiv:2410.14111•ACM/IEEE Design Automation Conference (DAC)•Jun 23, 2024•7 citations

Quantum approximate optimization for combinatorial problems with constraints

Yue Ruan, Zhiqiang Yuan, Xiling Xue +1 more

Information Sciences•Nov 11, 2022•24 citations

Solving Combinatorial Optimization Problem Over Graph Through QUBO Transformation and Deep Reinforcement Learning

Tianle Pu, Chao Chen, Li Zeng +3 more

Many complex problems encountered in both production and daily life can be conceptualized as combinatorial optimization problems (COPs). Many ad-hoc deep learning methods have been proposed to solve these problems, but there still lacks an effective unified framework. In this work, we take a step towards this goal by designing an unified end-to-end deep reinforcement learning framework named CONQUER. CONQUER first translates various COPs into the unified QUBO (Quadratic Unconstrained Binary Optimization) formalization, and afterwards trains a neural QUBO solver to search high-quality strategies, which are then checked and repaired to obtain the solutions for original COPs. The QUBO solver is a purely data-driven neural model, which does not rely on any expert inputs. We adopt the graph transformer network to represent the QUBO graph, so as to effectively capture the features of problem objective and constraints. The Deep Q-network is utilized to train the QUBO solver to find long-sighted solution strategies. Despite being trained solely on small synthetic graphs, CONQUER exhibits promising generality on much larger instances, and can be easily adapted to real-world datasets. Experimentally, we show that CONQUER achieves the state-of-the-art performances on four COPs over graphs, including minimum vertex cover, maximum independent set, maximum clique and maximum cut problem.

Proceedings•Dec 9, 2024•2 citations

Deriving Compact QUBO Models via Multilevel Constraint Transformation

Оксана Сергеевна Пичугина, Yingcong Tan, J. Christopher Beck

With the advances in customized hardware for quantum annealing and digital/CMOS Annealing, Quadratic Unconstrained Binary Optimization (QUBO) models have received growing attention in the optimization literature. Motivated by an existing general-purpose approach that derives QUBO models from binary linear programs (BLP), we propose a novel Multilevel Constraint Transformation Scheme (MLCTS) that derives QUBO models with fewer ancillary binary variables. We formulate sufficient conditions for the existence of a compact QUBO formulation (i.e., in the original BLP decision space) in terms of constraint levelness and demonstrate the flexibility and applicability of MLCTS on synthetic examples and several well-known combinatorial optimization problems, i.e., the Maximum 2-Satisfiability Problem, the Linear Ordering Problem, the Community Detection Problem, and the Maximum Independence Set Problem. For a proof-of-concept, we compare the performance of two QUBO models for the latter problem on both a general-purpose software-based solver and a hardware-based QUBO solver. The MLCTS-derived models demonstrate significantly better performance for both solvers, in particular, solving up to seven times more instances with the hardware-based approach.

arXiv:2404.03610•Apr 4, 2024•0 citations

QUBO: Quadratic Unconstrained Binary Optimization Problem

Gary A. Kochenberger, Fred Glover, Haibo Wang

Springer eBooks•Dec 17, 2024•1 citation
Computationally challenging combinatorial optimization problems (COPs) play a fundamental role in various applications. To tackle COPs, many Ising machines and Quadratic Unconstrained Binary Optimization (QUBO) solvers have been proposed, which typically involve direct transformation of COPs into Ising models or equivalent QUBO forms (D-QUBO). However, when addressing COPs with inequality constraints, this D-QUBO approach introduces numerous extra auxiliary variables, resulting in a substantially larger search space, increased hardware costs, and reduced solving efficiency. In this work, we propose HyCiM, a novel hybrid computing-inmemory (CiM) based QUBO solver framework, designed to overcome aforementioned challenges. The proposed framework consists of (i) an innovative transformation method (first to our known) that converts COPs with inequality constraints into an inequality-QUBO form, thus eliminating the need of expensive auxiliary variables and associated calculations; (ii) "inequality filter", a ferroelectric FET (FeFET)-based CiM circuit that accelerates the inequality evaluation, and filters out infeasible input configurations; (iii) a FeFET-based CiM annealer that is capable of approaching global solutions of COPs via iterative QUBO computations within a simulated annealing process. The evaluation results show that HyCiM drastically narrows down the search space, eliminating 2100 to 22536 infeasible input configurations compared to the conventional D-QUBO approach. Consequently, the narrowed search space, reduced to 2100 feasible input configurations, leads to a substantial hardware area overhead reduction, ranging from 88.06% to 99.96%. Additionally, HyCiM consistently exhibits a high solving efficiency, achieving a remarkable average success rate of 98.54%, whereas D-QUBO implementatoin shows only 10.75%.
Many complex problems encountered in both production and daily life can be conceptualized as combinatorial optimization problems (COPs). Many ad-hoc deep learning methods have been proposed to solve these problems, but there still lacks an effective unified framework. In this work, we take a step towards this goal by designing an unified end-to-end deep reinforcement learning framework named CONQUER. CONQUER first translates various COPs into the unified QUBO (Quadratic Unconstrained Binary Optimization) formalization, and afterwards trains a neural QUBO solver to search high-quality strategies, which are then checked and repaired to obtain the solutions for original COPs. The QUBO solver is a purely data-driven neural model, which does not rely on any expert inputs. We adopt the graph transformer network to represent the QUBO graph, so as to effectively capture the features of problem objective and constraints. The Deep Q-network is utilized to train the QUBO solver to find long-sighted solution strategies. Despite being trained solely on small synthetic graphs, CONQUER exhibits promising generality on much larger instances, and can be easily adapted to real-world datasets. Experimentally, we show that CONQUER achieves the state-of-the-art performances on four COPs over graphs, including minimum vertex cover, maximum independent set, maximum clique and maximum cut problem.
With the advances in customized hardware for quantum annealing and digital/CMOS Annealing, Quadratic Unconstrained Binary Optimization (QUBO) models have received growing attention in the optimization literature. Motivated by an existing general-purpose approach that derives QUBO models from binary linear programs (BLP), we propose a novel Multilevel Constraint Transformation Scheme (MLCTS) that derives QUBO models with fewer ancillary binary variables. We formulate sufficient conditions for the existence of a compact QUBO formulation (i.e., in the original BLP decision space) in terms of constraint levelness and demonstrate the flexibility and applicability of MLCTS on synthetic examples and several well-known combinatorial optimization problems, i.e., the Maximum 2-Satisfiability Problem, the Linear Ordering Problem, the Community Detection Problem, and the Maximum Independence Set Problem. For a proof-of-concept, we compare the performance of two QUBO models for the latter problem on both a general-purpose software-based solver and a hardware-based QUBO solver. The MLCTS-derived models demonstrate significantly better performance for both solvers, in particular, solving up to seven times more instances with the hardware-based approach.