Empirical risk minimization: probabilistic complexity and stepsize strategy

Chin Pang Ho () and Panos Parpas
Additional contact information
Chin Pang Ho: Imperial College London
Panos Parpas: Imperial College London

Computational Optimization and Applications, 2019, vol. 73, issue 2, No 2, 387-410

Abstract: Abstract Empirical risk minimization is recognized as a special form in standard convex optimization. When using a first order method, the Lipschitz constant of the empirical risk plays a crucial role in the convergence analysis and stepsize strategies for these problems. We derive the probabilistic bounds for such Lipschitz constants using random matrix theory. We show that, on average, the Lipschitz constant is bounded by the ratio of the dimension of the problem to the amount of training data. We use our results to develop a new stepsize strategy for first order methods. The proposed algorithm, Probabilistic Upper-bound Guided stepsize strategy, outperforms the regular stepsize strategies with strong theoretical guarantee on its performance.

Keywords: Empirical risk minimization; Complexity analysis; Stepsize strategy (search for similar items in EconPapers)
Date: 2019
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10589-019-00080-2 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:coopap:v:73:y:2019:i:2:d:10.1007_s10589-019-00080-2

Ordering information: This journal article can be ordered from
http://www.springer.com/math/journal/10589

DOI: 10.1007/s10589-019-00080-2

Access Statistics for this article

Computational Optimization and Applications is currently edited by William W. Hager

More articles in Computational Optimization and Applications from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().