Privately Answering Classification Queries in the Agnostic PAC Model

Anupama Nandi, Raef Bassily
Proceedings of the 31st International Conference on Algorithmic Learning Theory, PMLR 117:687-703, 2020.

Abstract

We revisit the problem of differentially private release of classification queries. In this problem, the goal is to design an algorithm that can accurately answer a sequence of classification queries based on a private training set while ensuring differential privacy. We formally study this problem in the agnostic PAC model and derive a new upper bound on the private sample complexity. Our results improve over those obtained in a recent work (Bassily et al., 2018) for the agnostic PAC setting. In particular, we give an improved construction that yields a tighter upper bound on the sample complexity. Moreover, unlike (Bassily et al., 2018), our accuracy guarantee does not involve any blow-up in the approximation error associated with the given hypothesis class. Given any hypothesis class with VC-dimension $d$, we show that our construction can privately answer up to $m$ classification queries with average excess error $\alpha$ using a private sample of size $\approx \frac{d}{\alpha^2}\,\max\left(1, \sqrt{m}\,\alpha^{3/2}\right)$. Using recent results on private learning with auxiliary public data, we extend our construction to show that one can privately answer any number of classification queries with average excess error $\alpha$ using a private sample of size $\approx \frac{d}{\alpha^2}\,\max\left(1, \sqrt{d}\,\alpha\right)$. When $\alpha=O\left(\frac{1}{\sqrt{d}}\right)$, our private sample complexity bound is essentially optimal.

Cite this Paper


BibTeX
@InProceedings{pmlr-v117-nandi20a, title = {Privately Answering Classification Queries in the Agnostic PAC Model}, author = {Nandi, Anupama and Bassily, Raef}, booktitle = {Proceedings of the 31st International Conference on Algorithmic Learning Theory}, pages = {687--703}, year = {2020}, editor = {Kontorovich, Aryeh and Neu, Gergely}, volume = {117}, series = {Proceedings of Machine Learning Research}, month = {08 Feb--11 Feb}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v117/nandi20a/nandi20a.pdf}, url = {https://proceedings.mlr.press/v117/nandi20a.html}, abstract = {We revisit the problem of differentially private release of classification queries. In this problem, the goal is to design an algorithm that can accurately answer a sequence of classification queries based on a private training set while ensuring differential privacy. We formally study this problem in the agnostic PAC model and derive a new upper bound on the private sample complexity. Our results improve over those obtained in a recent work (Bassily et al., 2018) for the agnostic PAC setting. In particular, we give an improved construction that yields a tighter upper bound on the sample complexity. Moreover, unlike (Bassily et al., 2018), our accuracy guarantee does not involve any blow-up in the approximation error associated with the given hypothesis class. Given any hypothesis class with VC-dimension $d$, we show that our construction can privately answer up to $m$ classification queries with average excess error $\alpha$ using a private sample of size $\approx \frac{d}{\alpha^2}\,\max\left(1, \sqrt{m}\,\alpha^{3/2}\right)$. Using recent results on private learning with auxiliary public data, we extend our construction to show that one can privately answer any number of classification queries with average excess error $\alpha$ using a private sample of size $\approx \frac{d}{\alpha^2}\,\max\left(1, \sqrt{d}\,\alpha\right)$. When $\alpha=O\left(\frac{1}{\sqrt{d}}\right)$, our private sample complexity bound is essentially optimal.} }
Endnote
%0 Conference Paper %T Privately Answering Classification Queries in the Agnostic PAC Model %A Anupama Nandi %A Raef Bassily %B Proceedings of the 31st International Conference on Algorithmic Learning Theory %C Proceedings of Machine Learning Research %D 2020 %E Aryeh Kontorovich %E Gergely Neu %F pmlr-v117-nandi20a %I PMLR %P 687--703 %U https://proceedings.mlr.press/v117/nandi20a.html %V 117 %X We revisit the problem of differentially private release of classification queries. In this problem, the goal is to design an algorithm that can accurately answer a sequence of classification queries based on a private training set while ensuring differential privacy. We formally study this problem in the agnostic PAC model and derive a new upper bound on the private sample complexity. Our results improve over those obtained in a recent work (Bassily et al., 2018) for the agnostic PAC setting. In particular, we give an improved construction that yields a tighter upper bound on the sample complexity. Moreover, unlike (Bassily et al., 2018), our accuracy guarantee does not involve any blow-up in the approximation error associated with the given hypothesis class. Given any hypothesis class with VC-dimension $d$, we show that our construction can privately answer up to $m$ classification queries with average excess error $\alpha$ using a private sample of size $\approx \frac{d}{\alpha^2}\,\max\left(1, \sqrt{m}\,\alpha^{3/2}\right)$. Using recent results on private learning with auxiliary public data, we extend our construction to show that one can privately answer any number of classification queries with average excess error $\alpha$ using a private sample of size $\approx \frac{d}{\alpha^2}\,\max\left(1, \sqrt{d}\,\alpha\right)$. When $\alpha=O\left(\frac{1}{\sqrt{d}}\right)$, our private sample complexity bound is essentially optimal.
APA
Nandi, A. & Bassily, R.. (2020). Privately Answering Classification Queries in the Agnostic PAC Model. Proceedings of the 31st International Conference on Algorithmic Learning Theory, in Proceedings of Machine Learning Research 117:687-703 Available from https://proceedings.mlr.press/v117/nandi20a.html.

Related Material