Robust deep semi-supervised learning with label propagation and differential privacy

Frontiers in Computer Science

Computer Science

Shenghong Li¹  Zhicong Yan¹  Zhongli Duan²  Yuanyuan Zhao³ 

[1] Department of Cyber Science and Engineering, Shanghai Jiao Tong University, Shanghai, China;School of Art and Design, Zhengzhou Institute of Industrial Application Technology, Zhengzhou University, Zhengzhou, Hebei, China;School of Information Science and Engineering, Hangzhou Normal University, Hangzhou, Zhejiang, China;

关键词: deep semi-supervised learning;    label propagation;    differential privacy;    robust learning;    mixup data augmentation;   

DOI  :  10.3389/fcomp.2023.1114186

 received in 2022-12-02, accepted in 2023-04-20,  发布年份 2023

来源: Frontiers

【 摘 要 】

Semi-supervised learning (SSL) methods provide a powerful tool for utilizing abundant unlabeled data to strengthen standard supervised learning. Traditional graph-based SSL methods prevail in classical SSL problems for their intuitional implementation and effective performance. However, they encounter troubles when applying to image classification followed by modern deep learning, since the diffusion algorithms face the curse of dimensionality. In this study, we propose a simple and efficient SSL method, combining a graph-based SSL paradigm with differential privacy. We aim at developing coherent latent feature space of deep neural networks so that the diffusion algorithm in the latent space can give more precise predictions for unlabeled data. Our approach achieves state-of-the-art performance on the Cifar10, Cifar100, and Mini-imagenet benchmark datasets and obtains an error rate of 18.56% on Cifar10 using only 1% of all labels. Furthermore, our approach inherits the benefits of graph-based SSL methods with a simple training process and can be easily combined with any network architecture.

【 授权许可】

Unknown   
Copyright © 2023 Yan, Li, Duan and Zhao.