Quantum Continual Learning Overcoming Catastrophic Forgetting

• 作者: Wenjie Jiang Zhide Lu Dong-Ling Deng
• 来源期刊: 中国物理快报（英文版）
• 年，卷（期）: 2022,（5）
• 所属期刊栏目: GENERAL
• 页码范围: 26-38
• 页数: 13页
• 语种: 英文
• DOI: 10.1088/0256-307X/39/5/050303

摘要

Catastrophic forgetting describes the fact that machine learning models will likely forget the knowledge of previ-ously learned tasks after the learning process of a new one.It is a vital problem in the continual learning scenario and recently has attracted tremendous concern across different communities.We explore the catastrophic forget-ting phenomena in the context of quantum machine learning.It is found that,similar to those classical learning models based on neural networks,quantum learning systems likewise suffer from such forgetting problem in classi-fication tasks emerging from various application scenes.We show that based on the local geometrical information in the loss function landscape of the trained model,a uniform strategy can be adapted to overcome the forget-ting problem in the incremental learning setting.Our results uncover the catastrophic forgetting phenomena in quantum machine learning and offer a practical method to overcome this problem,which opens a new avenue for exploring potential quantum advantages towards continual learning.