Task-aware Privacy Preservation for Multi-dimensional Data

Jiangnan Cheng, Ao Tang, Sandeep Chinchali
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:3835-3851, 2022.

Abstract

Local differential privacy (LDP) can be adopted to anonymize richer user data attributes that will be input to sophisticated machine learning (ML) tasks. However, today’s LDP approaches are largely task-agnostic and often lead to severe performance loss – they simply inject noise to all data attributes according to a given privacy budget, regardless of what features are most relevant for the ultimate task. In this paper, we address how to significantly improve the ultimate task performance with multi-dimensional user data by considering a task-aware privacy preservation problem. The key idea is to use an encoder-decoder framework to learn (and anonymize) a task-relevant latent representation of user data. We obtain an analytical near-optimal solution for the linear setting with mean-squared error (MSE) task loss. We also provide an approximate solution through a gradient-based learning algorithm for general nonlinear cases. Extensive experiments demonstrate that our task-aware approach significantly improves ultimate task accuracy compared to standard benchmark LDP approaches with the same level of privacy guarantee.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-cheng22a, title = {Task-aware Privacy Preservation for Multi-dimensional Data}, author = {Cheng, Jiangnan and Tang, Ao and Chinchali, Sandeep}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {3835--3851}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/cheng22a/cheng22a.pdf}, url = {https://proceedings.mlr.press/v162/cheng22a.html}, abstract = {Local differential privacy (LDP) can be adopted to anonymize richer user data attributes that will be input to sophisticated machine learning (ML) tasks. However, today’s LDP approaches are largely task-agnostic and often lead to severe performance loss – they simply inject noise to all data attributes according to a given privacy budget, regardless of what features are most relevant for the ultimate task. In this paper, we address how to significantly improve the ultimate task performance with multi-dimensional user data by considering a task-aware privacy preservation problem. The key idea is to use an encoder-decoder framework to learn (and anonymize) a task-relevant latent representation of user data. We obtain an analytical near-optimal solution for the linear setting with mean-squared error (MSE) task loss. We also provide an approximate solution through a gradient-based learning algorithm for general nonlinear cases. Extensive experiments demonstrate that our task-aware approach significantly improves ultimate task accuracy compared to standard benchmark LDP approaches with the same level of privacy guarantee.} }
Endnote
%0 Conference Paper %T Task-aware Privacy Preservation for Multi-dimensional Data %A Jiangnan Cheng %A Ao Tang %A Sandeep Chinchali %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-cheng22a %I PMLR %P 3835--3851 %U https://proceedings.mlr.press/v162/cheng22a.html %V 162 %X Local differential privacy (LDP) can be adopted to anonymize richer user data attributes that will be input to sophisticated machine learning (ML) tasks. However, today’s LDP approaches are largely task-agnostic and often lead to severe performance loss – they simply inject noise to all data attributes according to a given privacy budget, regardless of what features are most relevant for the ultimate task. In this paper, we address how to significantly improve the ultimate task performance with multi-dimensional user data by considering a task-aware privacy preservation problem. The key idea is to use an encoder-decoder framework to learn (and anonymize) a task-relevant latent representation of user data. We obtain an analytical near-optimal solution for the linear setting with mean-squared error (MSE) task loss. We also provide an approximate solution through a gradient-based learning algorithm for general nonlinear cases. Extensive experiments demonstrate that our task-aware approach significantly improves ultimate task accuracy compared to standard benchmark LDP approaches with the same level of privacy guarantee.
APA
Cheng, J., Tang, A. & Chinchali, S.. (2022). Task-aware Privacy Preservation for Multi-dimensional Data. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:3835-3851 Available from https://proceedings.mlr.press/v162/cheng22a.html.

Related Material