Cookies?
Library Header Image
LSE Research Online LSE Library Services

Functional output regression with infimal convolution: exploring the Huber and ε-insensitive Losses

Lambert, Alex, Bouche, Dimitri, Szabo, Zoltan and d'Alché-Buc, Florence (2022) Functional output regression with infimal convolution: exploring the Huber and ε-insensitive Losses. In: International Conference on Machine Learning, 2022-07-17 - 2022-07-23, Baltimore, MD, United States.

[img] Text (Szabo_functional-output-regression-presentation-slides) - Other
Download (428kB)
[img] Text (Szabo_functional-output-regression-conference-paper) - Published Version
Download (2MB)

Abstract

The focus of the paper is functional output regression (FOR) with convoluted losses. While most existing work consider the square loss setting, we leverage extensions of the Huber and the ε-insensitive loss (induced by infimal convolution) and propose a flexible framework capable of handling various forms of outliers and sparsity in the FOR family. We derive computationally tractable algorithms relying on duality to tackle the resulting tasks in the context of vector-valued reproducing kernel Hilbert spaces. The efficiency of the approach is demonstrated and contrasted with the classical squared loss setting on both synthetic and real-world benchmarks.

Item Type: Conference or Workshop Item (Paper)
Official URL: https://proceedings.mlr.press/v162/lambert22a.html
Additional Information: © 2022 The Authors
Divisions: Statistics
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
H Social Sciences > HA Statistics
Date Deposited: 26 Jul 2022 09:24
Last Modified: 28 Jul 2022 11:09
URI: http://eprints.lse.ac.uk/id/eprint/115651

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics