Digital Annealer for quadratic unconstrained binary optimization: A comparative performance analysis

Şeker, Oylum; Tanoumand, Neda; Bodur, Merve

doi:10.1016/j.asoc.2022.109367

Cited by 13 publications

(4 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also plan to employ more weight setting methods like particle swarm optimization, ant colony optimization, or tabu search, to properly set weights of constraint and optimization terms for improving QAA performance. Furthermore, we plan to apply the QUBU formulas to developing algorithms for different computing machines that are similar to the quantum annealer, such as the digital annealer (DA) [61] and the coherent Ising machine (CIM) [62], to see if the DA or the CIM can obtain better solutions than the quantum annealer.…”

Section: Discussionmentioning

confidence: 99%

Classifying and Benchmarking Quantum Annealing Algorithms Based on Quadratic Unconstrained Binary Optimization for Solving NP-Hard Problems

Jiang,

Chu

2023

IEEE Access

View full text Add to dashboard Cite

Quantum annealing has the potential to outperform classical transistor-based computer technologies in tackling intricate combinatorial optimization problems. However, ongoing scientific debates cast doubts on whether quantum annealing devices (or quantum annealers) can genuinely provide better problem-solving capabilities than classical computers. The question of whether quantum annealing algorithms (QAAs) running on quantum annealers have computational advantages over classical algorithms (CAs) running on classical computers still remains unclear. This paper aims to clarify the question by classifying and benchmarking QAAs that utilize quadratic unconstrained binary optimization (QUBO) formulas to solve NP-hard problems. It proposes a four-class classification of QUBO formulas and exemplifies each class by QUBO formulas used by QAAs for solving specific NP-hard problems, such as the subset sum, maximum cut, vertex cover, 0/1 knapsack, graph coloring, Hamiltonian cycle, traveling salesperson, and job shop scheduling problems. The classification is based on the following two criteria: (i) Does the number of QUBO variables scale linearly with the problem input size? (ii) Does the QUBO formula have both the constraint term and the optimization term? QAAs are implemented and run on a D-Wave quantum annealer for benchmarking. They are benchmarked against related classical algorithms (CAs) running on classical computers in terms of the quality of the solution and the time to the solution. The benchmarking results reveal which classes of QUBO formulas are likely to provide advantages to QAAs over CAs. Furthermore, based on the benchmarking results, observations and suggestions are given for each class of QUBO formulas, facilitating the adoption of appropriate actions to improve the performance of QAAs.

show abstract

Section: Discussionmentioning

confidence: 99%

Classifying and Benchmarking Quantum Annealing Algorithms Based on Quadratic Unconstrained Binary Optimization for Solving NP-Hard Problems

Jiang,

Chu

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…SA-based or QA-based Ising machine studies have reported that the penalty method falls into a local minimum solution and fails to reach the optimum of the QUBO (eg., Refs. [21], [22]). Therefore, developing an efficient constraint-handling method is important.…”

Section: Optimum Optimummentioning

confidence: 99%

Spin-Variable Reduction Method for Handling Linear Equality Constraints in Ising Machines

Shirai

Togawa

2023

IEEE Trans. Comput.

View full text Add to dashboard Cite

We propose a spin-variable reduction method for Ising machines to handle linear equality constraints in a combinatorial optimization problem. Ising machines including quantum-annealing machines can effectively solve combinatorial optimization problems. They are designed to find the lowest-energy solution of a quadratic unconstrained binary optimization (QUBO), which is mapped from the combinatorial optimization problem. The proposed method reduces the number of binary variables to formulate the QUBO compared to the conventional penalty method. We demonstrate a sufficient condition to obtain the optimum of the combinatorial optimization problem in the spin-variable reduction method and its general applicability. We apply it to typical combinatorial optimization problems, such as the graph k-partitioning problem and the quadratic assignment problem. Experiments using simulated-annealing and quantum-annealing based Ising machines demonstrate that the spin-variable reduction method outperforms the penalty method. The proposed method extends the application of Ising machines to larger-size combinatorial optimization problems with linear equality constraints.

show abstract

“…To the best of our knowledge, this is the first study to tackle this challenging problem, which involves optimizing network throughput, cost, and resilience simultaneously. Traditional meta-heuristic [21][22][23] and exact [24] multi-objective optimization approaches are not practical in this setting due to their high running times, which make them impractical for online servicing of new traffic requests. In contrast, our approach utilizes RL to rapidly provide optimal RWA solutions, making it a practical and efficient solution for this problem.…”

Section: Introductionmentioning

confidence: 99%

Interpreting multi-objective reinforcement learning for routing and wavelength assignment in optical networks

Nallaperuma¹,

Gan²,

Nevin³

et al. 2023

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

Performance optimization literature in optical networks predominantly consists of single objective optimization studies while often in practice multiple performance goals are to be met. This study addresses this issue with a generalized reinforcement learning (RL) model for parameter optimization in optical networks in the presence of multiple performance goals. Using this generic model, two multi-objective variants of a classical optimization problem in optical network operation, routing and wavelength assignment (RWA), are derived and solved to near optimality. The allocated route and wavelength for each demand are optimized with respect to the number of accepted services, the number of transmitters, and network availability. The resultant approximated Pareto front provides a set of solutions from which network operators can make decisions based on their preferences for particular objectives. These results contribute to the understanding of the relationships between different network parameters and performance metrics, which would be beneficial in future network design and growth. Moreover, benchmarking results against the state-of-the-art RWA heuristics suggest the applicability of RL in dynamic settings under changing traffic and generalizability for unseen traffic.

show abstract

Digital Annealer for quadratic unconstrained binary optimization: A comparative performance analysis

Cited by 13 publications

References 52 publications

Classifying and Benchmarking Quantum Annealing Algorithms Based on Quadratic Unconstrained Binary Optimization for Solving NP-Hard Problems

Classifying and Benchmarking Quantum Annealing Algorithms Based on Quadratic Unconstrained Binary Optimization for Solving NP-Hard Problems

Spin-Variable Reduction Method for Handling Linear Equality Constraints in Ising Machines

Interpreting multi-objective reinforcement learning for routing and wavelength assignment in optical networks

Contact Info

Product

Resources

About