Privacy Enhancement for Cloud-Based Few-Shot Learning

被引:1
作者
Parnami, Archit [1 ]
Usama, Muhammad [1 ]
Fan, Liyue [1 ]
Lee, Minwoo [1 ]
机构
[1] Univ North Carolina Charlotte, Dept Comp Sci, Charlotte, NC 28223 USA
来源
2022 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN) | 2022年
基金
美国国家科学基金会;
关键词
few-shot learning; privacy; cloud; image classification; differential privacy; meta-learning;
D O I
10.1109/IJCNN55064.2022.9892135
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Requiring less data for accurate models, few-shot learning has shown robustness and generality in many application domains. However, deploying few-shot models in untrusted environments may inflict privacy concerns, e.g., attacks or adversaries that may breach the privacy of user-supplied data. This paper studies the privacy enhancement for the few-shot learning in an untrusted environment, e.g., the cloud, by establishing a novel privacy-preserved embedding space that preserves the privacy of data and maintains the accuracy of the model. We examine the impact of various image privacy methods such as blurring, pixelization, Gaussian noise, and differentially private pixelization (DP-Pix) on few-shot image classification and propose a method that learns privacy-preserved representation through the joint loss. The empirical results show how privacy-performance trade-off can be negotiated for privacy-enhanced few-shot learning.
引用
收藏
页数:10
相关论文
共 40 条
  • [1] A. Team, 2016, Conference on Predictive APIs and Apps, P1
  • [2] Deep Learning with Differential Privacy
    Abadi, Martin
    Chu, Andy
    Goodfellow, Ian
    McMahan, H. Brendan
    Mironov, Ilya
    Talwar, Kunal
    Zhang, Li
    [J]. CCS'16: PROCEEDINGS OF THE 2016 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY, 2016, : 308 - 318
  • [3] [Anonymous], 2019, Few-shot learning: A survey
  • [4] [Anonymous], 2021, FEW SHOT IMAGE CLASS
  • [5] Bagdasaryan E., 2019, Differential privacy has disparate impact on model accuracy
  • [6] Bisong E., 2019, Building Machine Learning and Deep Learning Models on Google Cloud Platform, P581
  • [7] Bridle JS., 1990, NATO ASI series, V68
  • [8] Cabrero-Holgueras Jose, 2021, Proceedings on Privacy Enhancing Technologies, V2021, P139, DOI 10.2478/popets-2021-0064
  • [9] Clark A., 2015, Pillow (pil fork) documentation, DOI DOI 10.5281/ZENODO.8104287
  • [10] The Algorithmic Foundations of Differential Privacy
    Dwork, Cynthia
    Roth, Aaron
    [J]. FOUNDATIONS AND TRENDS IN THEORETICAL COMPUTER SCIENCE, 2013, 9 (3-4): : 211 - 406