Overview
- Reviews the main approaches to problems of model selection and error estimation
- Simplifies most of the technical aspects focusing on the applicability of the approaches
- Presents the intuitions behind the methods, the formalism, and practical algorithms
Part of the book series: Modeling and Optimization in Science and Technologies (MOST, volume 15)
Access this book
Tax calculation will be finalised at checkout
Other ways to access
Table of contents (10 chapters)
Keywords
- Statistical Learning Theory
- Empirical Data
- Model Selection
- Error Estimation
- Resampling Methods
- Complexity-Based Methods
- Union and Shell Bounds
- Vapnik-Chernovenkis Theory
- Rademacher Complexity Theory
- Compression Bound
- Algorithmic Stability Theory
- PAC-Bayes Theory
- Differential Privacy Theory
- Data-driven models
About this book
How can we select the best performing data-driven model? How can we rigorously estimate its generalization error? Statistical learning theory answers these questions by deriving non-asymptotic bounds on the generalization error of a model or, in other words, by upper bounding the true error of the learned model based just on quantities computed on the available data. However, for a long time, Statistical learning theory has been considered only an abstract theoretical framework, useful for inspiring new learning approaches, but with limited applicability to practical problems. The purpose of this book is to give an intelligible overview of the problems of model selection and error estimation, by focusing on the ideas behind the different statistical learning theory approaches and simplifying most of the technical aspects with the purpose of making them more accessible and usable in practice. The book starts by presenting the seminal works of the 80’s and includes the most recent results. It discusses open problems and outlines future directions for research.
Authors and Affiliations
About the author
Bibliographic Information
Book Title: Model Selection and Error Estimation in a Nutshell
Authors: Luca Oneto
Series Title: Modeling and Optimization in Science and Technologies
DOI: https://doi.org/10.1007/978-3-030-24359-3
Publisher: Springer Cham
eBook Packages: Intelligent Technologies and Robotics, Intelligent Technologies and Robotics (R0)
Copyright Information: Springer Nature Switzerland AG 2020
Hardcover ISBN: 978-3-030-24358-6Published: 25 July 2019
Softcover ISBN: 978-3-030-24361-6Published: 14 August 2020
eBook ISBN: 978-3-030-24359-3Published: 17 July 2019
Series ISSN: 2196-7326
Series E-ISSN: 2196-7334
Edition Number: 1
Number of Pages: XIII, 132
Number of Illustrations: 62 b/w illustrations
Topics: Computational Intelligence, Statistical Theory and Methods, Data Mining and Knowledge Discovery