How to choose your loss when designing a Siamese Neural Network ? Contrastive, Triplet or Quadruplet ?

Medium post

Abstract

Deep Similarity Learning is the training of a deep learning architecture to learn to detect similarity and disimilarity between two inputs (or more). In this article, I presented, studied and compared three of the most popular losses for the task of training Siamese Network: contrastive, triplet and quadruplet loss.

Publication
How to choose your loss when designing a Siamese Neural Network ? Contrastive, Triplet or Quadruplet ?
Avatar
Thomas Di Martino
PhD Student in AI & Remote Sensing

My research interests include deep learning technologies, automatic feature extraction and computer vision, all of them applied to Remote Sensing problematics, more precisely to Synthetic Aperture Radar (SAR) acquisitions.

Related