Quantum annealing for semi-supervised learning

• 作者: Yu-Lin Zheng Wen Zhang Cheng Zhou Wei Geng
• 来源期刊: 中国物理B（英文版）
• 年，卷（期）: 2021,（4）
• 所属期刊栏目: SPECIAL TOPIC—Quantum computation and quantum simulation
• 页码范围: 80-87
• 页数: 8页
• 语种: 英文
• DOI: 10.1088/1674-1056/abe298

摘要

Recent advances in quantum technology have led to the development and the manufacturing of programmable quan-tum annealers that promise to solve certain combinatorial optimization problems faster than their classical counterparts.Semi-supervised learning is a machine learning technique that makes use of both labeled and unlabeled data for training,which enables a good classifier with only a small amount of labeled data.In this paper,we propose and theoretically an-alyze a graph-based semi-supervised learning method with the aid of the quantum annealing technique,which efficiently utilizes the quantum resources while maintaining good accuracy.We illustrate two classification examples,suggesting the feasibility of this method even with a small portion(30％)of labeled data involved.