NoPeek: Information leakage reduction to share activations in distributed deep learning
Name
2008.09161.pdf
Description
Submitted version
Size
13.07 MB
Format
Unknown
Checksum (MD5)
b9dfb9e7f69fbe7582d6c5a042a575fa
Author(s)
Vepakomma, Praneeth
Singh, Abhishek
Gupta, Otkrist
Raskar, Ramesh
Date Issued
2020
Journal
IEEE International Conference on Data Mining Workshops, ICDMW
Publisher
IEEE
Citation
Vepakomma, P, Singh, A, Gupta, O and Raskar, R. "NoPeek: Information leakage reduction to share activations in distributed deep learning." IEEE International Conference on Data Mining Workshops, ICDMW, 2020-November.
Version
Original manuscript
Abstract
For distributed machine learning with sensitive data, we demonstrate how minimizing distance correlation between raw data and intermediary representations reduces leakage of sensitive raw data patterns across client communications while maintaining model accuracy. Leakage (measured using distance correlation between input and intermediate representations) is the risk associated with the invertibility of raw data from intermediary representations. This can prevent client entities that hold sensitive data from using distributed deep learning services. We demonstrate that our method is resilient to such reconstruction attacks and is based on reduction of distance correlation between raw data and learned representations during training and inference with image datasets. We prevent such reconstruction of raw data while maintaining information required to sustain good classification accuracies.
MIT Department
Program in Media Arts and Sciences (Massachusetts Institute of Technology)
Massachusetts Institute of Technology. Media Laboratory
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1109/ICDMW51313.2020.00134
