Noise in Classification
This work tackles the problem of robust classification under noise for machine learning practitioners, but it appears incremental as it reviews existing approaches rather than introducing new methods.
The chapter addresses the challenge of learning linear thresholds with adversarial noise, which is notoriously hard in worst-case scenarios, and explores approaches to overcome these difficulties by leveraging natural assumptions on data generation.
This chapter considers the computational and statistical aspects of learning linear thresholds in presence of noise. When there is no noise, several algorithms exist that efficiently learn near-optimal linear thresholds using a small amount of data. However, even a small amount of adversarial noise makes this problem notoriously hard in the worst-case. We discuss approaches for dealing with these negative results by exploiting natural assumptions on the data-generating process.