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

Representation learning has significantly been developed with the advance ofcontrastive learning methods. Most of those methods have benefited from variousdata augmentations that are carefully designated to maintain their identitiesso that the images transformed from the same instance can still be retrieved.However, those carefully designed transformations limited us to further explorethe novel patterns exposed by other transformations. Meanwhile, as found in ourexperiments, the strong augmentations distorted the images' structures,resulting in difficult retrieval. Thus, we propose a general framework calledContrastive Learning with Stronger Augmentations~(CLSA) to complement currentcontrastive learning approaches. Here, the distribution divergence between theweakly and strongly augmented images over the representation bank is adopted tosupervise the retrieval of strongly augmented queries from a pool of instances.Experiments on the ImageNet dataset and downstream datasets showed theinformation from the strongly augmented images can significantly boost theperformance. For example, CLSA achieves top-1 accuracy of 76.2% on ImageNetwith a standard ResNet-50 architecture with a single-layer classifierfine-tuned, which is almost the same level as 76.5% of supervised results. Thecode and pre-trained models are available inhttps://github.com/maple-research-lab/CLSA.