Robust Sparse Regression under Adversarial Corruption

By Yudong Chen, Constantine Caramanis, Shie Mannor. 2013

140 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning (2018, methodology): “This is one of the main difference from prior work [12], [17], which assume the knowledge of the mean and covariance of the legitimate data.”
Distributed Robust Learning (2014, methodology): “Here we refer the readers to [5] for more details about Algorithm 5.”
Robust Regression via Hard Thresholding (2015, background): “In contrast, [6] studied RLSR with less stringent assumptions, allowing arbitrary corruptions in response variables as well as in the data matrix X , and proposed a trimmed inner product based algorithm for the problem.”
Robust Logistic Regression and Classification (2014, methodology): “Based on the above concentration property, we can obtain following bound on the magnitude of a collection of sub-Gaussian random variables [3].”
Learning with Bad Training Data via Iterative Trimmed Loss Minimization (2018, background): “Another line of recent work on robust regression consider strong robustness where the adversary poisons both the inputs and outputs, in both low-dimensional [DKK + 18, PSBR18, KKM18] and high dimensional [CCM13, BDLS17, LSLC18] settings.”
Consistent Robust Regression (2017, methodology): “We note that although hard thresholding techniques have been applied to the RLSR problem earlier [3, 6], none of those methods are able to guarantee a consistent solution to the problem.”
Efficient Algorithms for Outlier-Robust Regression (2018, background): “, Herman and Strohmer (2010)) (although there are some contexts where both the covariates (the x’s) and labels may be subject to a small adversarial corruption (Chen et al., 2013)).”
On the Practicality of Integrity Attacks on Document-Level Sentiment Analysis (2014, methodology): “Numerous techniques have been designed to deal with the noise and errors present in benign environments for sentiment analysis.”
Choosing the Sample with Lowest Loss makes SGD Robust (2020, background): “Another line of recent work considers robustness in the high-dimensional setting ([27, 36, 8, 3, 24]) In this version, our focus is on general stochastic optimization in presence of outliers.”