On the Adversarial Robustness of LASSO Based Feature Selection

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.

References in corpus (14)

• Explaining and Harnessing Adversarial Examples
• Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
• Evasion Attacks against Machine Learning at Test Time
• Task-Driven Dictionary Learning
• High-dimensional regression with noisy and missing data: Provable guarantees with nonconvexity
• Is feature selection secure against training data poisoning?
• Group-Sparse Signal Denoising: Non-Convex Regularization, Convex Optimization
• Dynamic Compressive Sensing of Time-Varying Signals via Approximate Message Passing
• Structure-Aware Bayesian Compressive Sensing for Frequency-Hopping Spectrum Estimation with Missing Observations
• Adversarial Attacks on Machine Learning Systems for High-Frequency Trading
• Robust Learning of Fixed-Structure Bayesian Networks
• On the Adversarial Robustness of Subspace Learning
• On Security and Sparsity of Linear Classifiers for Adversarial Settings
• Adversarial Robustness of Supervised Sparse Coding

Cited by in corpus (1)

• Provably Efficient Black-Box Action Poisoning Attacks Against Reinforcement Learning