Cross-Problem Parameter Transfer in Quantum Approximate Optimization Algorithm: A Machine Learning Approach

2025-04-14

AI Summary

Quantum Approximate Optimization Algorithm (QAOA) suffers from costly parameter optimization for each new NP-hard combinatorial problem, limiting its practicality. Method: This work systematically investigates cross-problem parameter transferability—specifically, whether MaxCut-pretrained QAOA parameters can be directly reused or warm-started for the Maximum Independent Set (MIS) problem. We propose a machine learning–driven framework for inter-problem parameter selection, integrating supervised and unsupervised models with novel parameter similarity metrics to identify high-quality donor parameters, thereby extending transfer beyond isomorphic problems. Contribution/Results: Experiments demonstrate that our approach reduces the number of variational optimization iterations for MIS by up to 60% while preserving solution quality—achieving an average approximation ratio error below 2%. This establishes a new paradigm for QAOA parameter reuse, enhancing both the practical deployment and generalization capability of quantum-inspired heuristics.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Metareasoning and Metaheuristics
• Constraint Satisfaction and Optimization: Constraint Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Quantum Approximate Optimization Algorithm (QAOA) is one of the most promising candidates to achieve the quantum advantage in solving combinatorial optimization problems. The process of finding a good set of variational parameters in the QAOA circuit has proven to be challenging due to multiple factors, such as barren plateaus. As a result, there is growing interest in exploiting parameter transferability, where parameter sets optimized for one problem instance are transferred to another that could be more complex either to estimate the solution or to serve as a warm start for further optimization. But can we transfer parameters from one class of problems to another? Leveraging parameter sets learned from a well-studied class of problems could help navigate the less studied one, reducing optimization overhead and mitigating performance pitfalls. In this paper, we study whether pretrained QAOA parameters of MaxCut can be used as is or to warm start the Maximum Independent Set (MIS) circuits. Specifically, we design machine learning models to find good donor candidates optimized on MaxCut and apply their parameters to MIS acceptors. Our experimental results show that such parameter transfer can significantly reduce the number of optimization iterations required while achieving comparable approximation ratios.

Problem

• Transfer QAOA parameters between different optimization problems
• Reduce optimization overhead using pretrained MaxCut parameters for MIS
• Improve QAOA efficiency via machine learning-based parameter transfer

Innovation

• Transfer QAOA parameters across problem classes
• Use machine learning to select donor parameters
• Reduce optimization iterations with transferred parameters