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

Generalization of machine learning models trained on a set of source domainson unseen target domains with different statistics, is a challenging problem.While many approaches have been proposed to solve this problem, they onlyutilize source data during training but do not take advantage of the fact thata single target example is available at the time of inference. Motivated bythis, we propose a method that effectively uses the target sample duringinference beyond mere classification. Our method has three components - (i) Alabel-preserving feature or metric transformation on source data such that thesource samples are clustered in accordance with their class irrespective oftheir domain (ii) A generative model trained on the these features (iii) Alabel-preserving projection of the target point on the source-feature manifoldduring inference via solving an optimization problem on the input space of thegenerative model using the learned metric. Finally, the projected target isused in the classifier. Since the projected target feature comes from thesource manifold and has the same label as the real target by design, theclassifier is expected to perform better on it than the true target. Wedemonstrate that our method outperforms the state-of-the-art DomainGeneralization methods on multiple datasets and tasks.

Domain Generalization via Inference-time Label-Preserving Target Projections