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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

To ease the burden of labeling, unsupervised domain adaptation (UDA) aims totransfer knowledge in previous and related labeled datasets (sources) to a newunlabeled dataset (target). Despite impressive progress, prior methods alwaysneed to access the raw source data and develop data-dependent alignmentapproaches to recognize the target samples in a transductive learning manner,which may raise privacy concerns from source individuals. Several recentstudies resort to an alternative solution by exploiting the well-trainedwhite-box model from the source domain, yet, it may still leak the raw datathrough generative adversarial learning. This paper studies a practical andinteresting setting for UDA, where only black-box source models (i.e., onlynetwork predictions are available) are provided during adaptation in the targetdomain. To solve this problem, we propose a new two-step knowledge adaptationframework called DIstill and fine-tuNE (DINE). Taking into consideration thetarget data structure, DINE first distills the knowledge from the sourcepredictor to a customized target model, then fine-tunes the distilled model tofurther fit the target domain. Besides, neural networks are not required to beidentical across domains in DINE, even allowing effective adaptation on alow-resource device. Empirical results on three UDA scenarios (i.e.,single-source, multi-source, and partial-set) confirm that DINE achieves highlycompetitive performance compared to state-of-the-art data-dependent approaches.Code is available at \url{https://github.com/tim-learn/DINE/}.

DINE: Domain Adaptation from Single and Multiple Black-box Predictors