English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Few-shot learning is a challenging area of research that aims to learn newconcepts with only a few labeled samples of data. Recent works based onmetric-learning approaches leverage the meta-learning approach, which isencompassed by episodic tasks that make use a support (training) and query set(test) with the objective of learning a similarity comparison metric betweenthose sets. Due to the lack of data, the learning process of the embeddingnetwork becomes an important part of the few-shot task. Previous works haveaddressed this problem using metric learning approaches, but the properties ofthe underlying latent space and the separability of the difference classes onit was not entirely enforced. In this work, we propose two different lossfunctions which consider the importance of the embedding vectors by looking atthe intra-class and inter-class distance between the few data. The first lossfunction is the Proto-Triplet Loss, which is based on the original triplet losswith the modifications needed to better work on few-shot scenarios. The secondloss function, which we dub ICNN loss is based on an inter and intra classnearest neighbors score, which help us to assess the quality of embeddingsobtained from the trained network. Our results, obtained from a extensiveexperimental setup show a significant improvement in accuracy in theminiImagenNet benchmark compared to other metric-based few-shot learningmethods by a margin of 2%, demonstrating the capability of these loss functionsto allow the network to generalize better to previously unseen classes. In ourexperiments, we demonstrate competitive generalization capabilities to otherdomains, such as the Caltech CUB, Dogs and Cars datasets compared with thestate of the art.

SuSana Distancia is all you need: Enforcing class separability in metric  learning via two novel distance-based loss functions for few-shot image  classification