On the Adversarial Robustness of LASSO Based Feature Selection

TL;DR

This paper explores the vulnerability of LASSO-based feature selection to adversarial attacks, proposing a bi-level optimization approach to craft modifications that manipulate feature selection outcomes.

Contribution

It introduces a novel bi-level optimization framework for adversarial attacks on LASSO feature selection, addressing non-differentiability via reformulation and extending to related methods.

Findings

The proposed method effectively manipulates feature selection in synthetic data.

It demonstrates efficiency and effectiveness on real datasets.

The approach can be extended to group LASSO and sparse group LASSO.

Abstract

In this paper, we investigate the adversarial robustness of feature selection based on the $\ell_1$ regularized linear regression model, namely LASSO. In the considered model, there is a malicious adversary who can observe the whole dataset, and then will carefully modify the response values or the feature matrix in order to manipulate the selected features. We formulate the modification strategy of the adversary as a bi-level optimization problem. Due to the difficulty of the non-differentiability of the $\ell_1$ norm at the zero point, we reformulate the $\ell_1$ norm regularizer as linear inequality constraints. We employ the interior-point method to solve this reformulated LASSO problem and obtain the gradient information. Then we use the projected gradient descent method to design the modification strategy. In addition, We demonstrate that this method can be extended to other $\ell_1$ based feature selection methods, such as group LASSO and sparse group LASSO. Numerical examples with synthetic and real data illustrate that our method is efficient and effective.