Rethinking Positive Pairs in Contrastive Learning

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wu, Jiantao, Atito, Sara, Feng, Zhenhua, Mo, Shentong, Kitler, Josef, Awais, Muhammad
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913865625763840
author Wu, Jiantao
Atito, Sara
Feng, Zhenhua
Mo, Shentong
Kitler, Josef
Awais, Muhammad
author_facet Wu, Jiantao
Atito, Sara
Feng, Zhenhua
Mo, Shentong
Kitler, Josef
Awais, Muhammad
contents The training methods in AI do involve semantically distinct pairs of samples. However, their role typically is to enhance the between class separability. The actual notion of similarity is normally learned from semantically identical pairs. This paper presents SimLAP: a simple framework for learning visual representation from arbitrary pairs. SimLAP explores the possibility of learning similarity from semantically distinct sample pairs. The approach is motivated by the observation that for any pair of classes there exists a subspace in which semantically distinct samples exhibit similarity. This phenomenon can be exploited for a novel method of learning, which optimises the similarity of an arbitrary pair of samples, while simultaneously learning the enabling subspace. The feasibility of the approach will be demonstrated experimentally and its merits discussed.
format Preprint
id arxiv_https___arxiv_org_abs_2410_18200
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Rethinking Positive Pairs in Contrastive Learning
Wu, Jiantao
Atito, Sara
Feng, Zhenhua
Mo, Shentong
Kitler, Josef
Awais, Muhammad
Computer Vision and Pattern Recognition
Machine Learning
The training methods in AI do involve semantically distinct pairs of samples. However, their role typically is to enhance the between class separability. The actual notion of similarity is normally learned from semantically identical pairs. This paper presents SimLAP: a simple framework for learning visual representation from arbitrary pairs. SimLAP explores the possibility of learning similarity from semantically distinct sample pairs. The approach is motivated by the observation that for any pair of classes there exists a subspace in which semantically distinct samples exhibit similarity. This phenomenon can be exploited for a novel method of learning, which optimises the similarity of an arbitrary pair of samples, while simultaneously learning the enabling subspace. The feasibility of the approach will be demonstrated experimentally and its merits discussed.
title Rethinking Positive Pairs in Contrastive Learning
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2410.18200