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

Meta-learning has emerged as an efficient approach for constructing targetmodels based on support sets. For example, the meta-learned embeddings enablethe construction of target nearest-neighbor classifiers for specific tasks bypulling instances closer to their same-class neighbors. However, a singleinstance can be annotated from various latent attributes, making visuallysimilar instances inside or across support sets have different labels anddiverse relationships with others. Consequently, a uniform meta-learnedstrategy for inferring the target model from the support set fails to capturethe instance-wise ambiguous similarity. To this end, we propose Learning toDecompose Network (LeadNet) to contextualize the meta-learned``support-to-target'' strategy, leveraging the context of instances with one ormixed latent attributes in a support set. In particular, the comparisonrelationship between instances is decomposed w.r.t. multiple embedding spaces.LeadNet learns to automatically select the strategy associated with the rightattribute via incorporating the change of comparison across contexts} withpolysemous embeddings. We demonstrate the superiority of LeadNet in variousapplications, including exploring multiple views of confusing data,out-of-distribution recognition, and few-shot image classification.

Contextualizing Meta-Learning via Learning to Decompose