Minimax M-estimation under Adversarial Contamination

Sujay Bhatt, Guanhua Fang, Ping Li, Gennady Samorodnitsky
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:1906-1924, 2022.

Abstract

We present a new finite-sample analysis of Catoni’s M-estimator under adversarial contamination, where an adversary is allowed to corrupt a fraction of the samples arbitrarily. We make minimal assumptions on the distribution of the uncontaminated random variables, namely, we only assume the existence of a known upper bound $\upsilon_{\varepsilon} > 0$ on the $(1+\varepsilon)^{th}$ central moment of the random variables, namely, for $\varepsilon \in (0,1]$ \[ \mathbb{E}_{X_1 \sim \mathcal{D}} \Big| X_1 - \mu \Big|^{1+\varepsilon} \leq \upsilon_{\varepsilon}. \]{We} provide a lower bound on the minimax error rate for the mean estimation problem under adversarial corruption under this weak assumption, and establish that the proposed M-estimator achieves this lower bound (up to multiplicative constants). When the variance is infinite, the tolerance to contamination of any estimator reduces as $\varepsilon \downarrow 0$. We establish a tight upper bound that characterizes this bargain. To illustrate the usefulness of the derived robust M-estimator in an online setting, we present a bandit algorithm for the partially identifiable best arm identification problem that improves upon the sample complexity of the state of the art algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-bhatt22a, title = {Minimax M-estimation under Adversarial Contamination}, author = {Bhatt, Sujay and Fang, Guanhua and Li, Ping and Samorodnitsky, Gennady}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {1906--1924}, 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/bhatt22a/bhatt22a.pdf}, url = {https://proceedings.mlr.press/v162/bhatt22a.html}, abstract = {We present a new finite-sample analysis of Catoni’s M-estimator under adversarial contamination, where an adversary is allowed to corrupt a fraction of the samples arbitrarily. We make minimal assumptions on the distribution of the uncontaminated random variables, namely, we only assume the existence of a known upper bound $\upsilon_{\varepsilon} > 0$ on the $(1+\varepsilon)^{th}$ central moment of the random variables, namely, for $\varepsilon \in (0,1]$ \[ \mathbb{E}_{X_1 \sim \mathcal{D}} \Big| X_1 - \mu \Big|^{1+\varepsilon} \leq \upsilon_{\varepsilon}. \]{We} provide a lower bound on the minimax error rate for the mean estimation problem under adversarial corruption under this weak assumption, and establish that the proposed M-estimator achieves this lower bound (up to multiplicative constants). When the variance is infinite, the tolerance to contamination of any estimator reduces as $\varepsilon \downarrow 0$. We establish a tight upper bound that characterizes this bargain. To illustrate the usefulness of the derived robust M-estimator in an online setting, we present a bandit algorithm for the partially identifiable best arm identification problem that improves upon the sample complexity of the state of the art algorithms.} }
Endnote
%0 Conference Paper %T Minimax M-estimation under Adversarial Contamination %A Sujay Bhatt %A Guanhua Fang %A Ping Li %A Gennady Samorodnitsky %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-bhatt22a %I PMLR %P 1906--1924 %U https://proceedings.mlr.press/v162/bhatt22a.html %V 162 %X We present a new finite-sample analysis of Catoni’s M-estimator under adversarial contamination, where an adversary is allowed to corrupt a fraction of the samples arbitrarily. We make minimal assumptions on the distribution of the uncontaminated random variables, namely, we only assume the existence of a known upper bound $\upsilon_{\varepsilon} > 0$ on the $(1+\varepsilon)^{th}$ central moment of the random variables, namely, for $\varepsilon \in (0,1]$ \[ \mathbb{E}_{X_1 \sim \mathcal{D}} \Big| X_1 - \mu \Big|^{1+\varepsilon} \leq \upsilon_{\varepsilon}. \]{We} provide a lower bound on the minimax error rate for the mean estimation problem under adversarial corruption under this weak assumption, and establish that the proposed M-estimator achieves this lower bound (up to multiplicative constants). When the variance is infinite, the tolerance to contamination of any estimator reduces as $\varepsilon \downarrow 0$. We establish a tight upper bound that characterizes this bargain. To illustrate the usefulness of the derived robust M-estimator in an online setting, we present a bandit algorithm for the partially identifiable best arm identification problem that improves upon the sample complexity of the state of the art algorithms.
APA
Bhatt, S., Fang, G., Li, P. & Samorodnitsky, G.. (2022). Minimax M-estimation under Adversarial Contamination. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:1906-1924 Available from https://proceedings.mlr.press/v162/bhatt22a.html.

Related Material