Skip to Main content Skip to Navigation
Conference papers

A Local Mixup to Prevent Manifold Intrusion

Raphael Baena 1, 2 Lucas Drumetz 1, 3 Vincent Gripon 1, 2 
2 Lab-STICC_2AI - Equipe Algorithm Architecture Interactions
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance : UMR6285
3 Lab-STICC_OSE - Equipe Observations Signal & Environnement
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance : UMR6285
Abstract : Deployed in the context of supervised learning, Mixup is a data-dependent regularization technique that consists in linearly interpolating input samples and associated outputs. It has been shown to improve accuracy when used to train on standard machine learning datasets. However, authors have pointed out that Mixup can produce out-of-distribution virtual samples and even contradictions in the augmented training set, potentially resulting in adversarial effects. In this paper, we introduce Local Mixup in which distant input samples are weighted down when computing the loss. In constrained settings we demonstrate that Local Mixup can create a trade-off between bias and variance, with the extreme cases reducing to vanilla training and classical Mixup. Using standardized computer vision benchmarks, we also show that Local Mixup can improve accuracy.
Complete list of metadata

https://hal-imt-atlantique.archives-ouvertes.fr/hal-03683702
Contributor : Raphael Baena Connect in order to contact the contributor
Submitted on : Tuesday, May 31, 2022 - 6:55:04 PM
Last modification on : Friday, August 5, 2022 - 2:54:52 PM

File

20220214042103_872102_1084.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03683702, version 1

Citation

Raphael Baena, Lucas Drumetz, Vincent Gripon. A Local Mixup to Prevent Manifold Intrusion. EUSIPCO 2022, Aug 2022, Belgrade, Serbia. ⟨hal-03683702⟩

Share

Metrics

Record views

11

Files downloads

4