Fairness, Accuracy, and Unreliable Data
It tackles reliability issues in machine learning for applications where data properties complicate learning, but it appears incremental as it builds on existing theoretical domains.
The thesis addresses the mismatch between classical learning theory and real-world data distributions, focusing on fairness, strategic classification, and algorithmic robustness to improve machine learning reliability.
This thesis investigates three areas targeted at improving the reliability of machine learning; fairness in machine learning, strategic classification, and algorithmic robustness. Each of these domains has special properties or structure that can complicate learning. A theme throughout this thesis is thinking about ways in which a `plain' empirical risk minimization algorithm will be misleading or ineffective because of a mis-match between classical learning theory assumptions and specific properties of some data distribution in the wild. Theoretical understanding in eachof these domains can help guide best practices and allow for the design of effective, reliable, and robust systems.