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

Contrastive learning, especially self-supervised contrastive learning (SSCL),has achieved great success in extracting powerful features from unlabeled data.In this work, we contribute to the theoretical understanding of SSCL anduncover its connection to the classic data visualization method, stochasticneighbor embedding (SNE), whose goal is to preserve pairwise distances. Fromthe perspective of preserving neighboring information, SSCL can be viewed as aspecial case of SNE with the input space pairwise similarities specified bydata augmentation. The established correspondence facilitates deepertheoretical understanding of learned features of SSCL, as well asmethodological guidelines for practical improvement. Specifically, through thelens of SNE, we provide novel analysis on domain-agnostic augmentations,implicit bias and robustness of learned features. To illustrate the practicaladvantage, we demonstrate that the modifications from SNE to $t$-SNE can alsobe adopted in the SSCL setting, achieving significant improvement in bothin-distribution and out-of-distribution generalization.

Your Contrastive Learning Is Secretly Doing Stochastic Neighbor Embedding