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

In this paper we present an improved approach to prototypical semi-supervisedlearning for computer vision, in the context of leveraging a frozen foundationmodel as the backbone of our neural network. As a general tool, we proposeparametric von-Mises Fisher Stochastic Neighbour Embedding (vMF-SNE) to createmappings with neural networks between high-dimensional latent spaces thatpreserve local structure. This enables us to pretrain the projection head ofour network using the high-quality embeddings of the foundation model withvMF-SNE. We also propose soft multi-view pseudolabels, where predictions acrossmultiple views are combined to provide a more reliable supervision signalcompared to a consistency or swapped assignment approach. We demonstrate thatthese ideas improve upon P}redicting View-Assignments with Support Samples(PAWS), a current state-of-the-art semi-supervised learning method, as well asRobust PAWS (RoPAWS), over a range of benchmarking datasets. We also introducesimple $k$-means prototype selection, a technique that provides superiorperformance to other unsupervised label selection approaches in this context.These changes improve upon PAWS by an average of +2.9% for CIFAR-10 and +5.7%for CIFAR-100 with four labels per class, and by +15.2% for DeepWeeds, aparticularly challenging dataset for semi-supervised learning. We also achievenew state-of-the-art results in semi-supervised learning in this small labelregime for CIFAR-10 - 95.8% (+0.7%) and CIFAR-100 - 76.6% (+12.0%).

Improved Prototypical Semi-Supervised Learning with Foundation Models:  Prototype Selection, Parametric vMF-SNE Pretraining and Multi-view  Pseudolabelling