Algorithmic High-Dimensional Robust Statistics

ISBN-13: 9781108837811 / Twarda / 2023 / 300 str.

San Diego) Kane Daniel M. (University of California

Opis

This book presents general principles and scalable methodologies to deal with adversarial outliers in high-dimensional datasets.

Recenzje

'This is a timely book on efficient algorithms for computing robust statistics from noisy data. It presents lucid intuitive descriptions of the algorithms as well as precise statements of results with rigorous proofs - a nice combination indeed. The topic has seen fundamental breakthroughs over the last few years and the authors are among the leading contributors. The reader will get a ringside view of the developments.' Ravi Kannan, Visiting Professor, Indian Institute of Science

Spis treści

1. Introduction to robust statistics; 2. Efficient high-dimensional robust mean estimation; 3. Algorithmic refinements in robust mean estimation; 4. Robust covariance estimation; 5. List-decodable learning; 6. Robust estimation via higher moments; 7. Robust supervised learning; 8. Information-computation tradeoffs in high-dimensional robust statistics; A. Mathematical background; References; Index.