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

Domain adaptation targets at knowledge acquisition and dissemination from alabeled source domain to an unlabeled target domain under distribution shift.Still, the common requirement of identical class space shared across domainshinders applications of domain adaptation to partial-set domains. Recentadvances show that deep pre-trained models of large scale endow rich knowledgeto tackle diverse downstream tasks of small scale. Thus, there is a strongincentive to adapt models from large-scale domains to small-scale domains. Thispaper introduces Partial Domain Adaptation (PDA), a learning paradigm thatrelaxes the identical class space assumption to that the source class spacesubsumes the target class space. First, we present a theoretical analysis ofpartial domain adaptation, which uncovers the importance of estimating thetransferable probability of each class and each instance across domains. Then,we propose Selective Adversarial Network (SAN and SAN++) with a bi-levelselection strategy and an adversarial adaptation mechanism. The bi-levelselection strategy up-weighs each class and each instance simultaneously forsource supervised training, target self-training, and source-target adversarialadaptation through the transferable probability estimated alternately by themodel. Experiments on standard partial-set datasets and more challenging taskswith superclasses show that SAN++ outperforms several domain adaptationmethods.

From Big to Small: Adaptive Learning to Partial-Set Domains