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/

Zendy Plus

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

Zendy Tools

Zendy Open

Learning the disentangled representation of interpretable generative factorsof data is one of the foundations to allow artificial intelligence to thinklike people. In this paper, we propose the analogical training strategy for theunsupervised disentangled representation learning in generative models. Theanalogy is one of the typical cognitive processes, and our proposed strategy isbased on the observation that sample pairs in which one is different from theother in one specific generative factor show the same analogical relation.Thus, the generator is trained to generate sample pairs from which a designedclassifier can identify the underlying analogical relation. In addition, wepropose a disentanglement metric called the subspace score, which is inspiredby subspace learning methods and does not require supervised information.Experiments show that our proposed training strategy allows the generativemodels to find the disentangled factors, and that our methods can givecompetitive performances as compared with the state-of-the-art methods.

Unsupervised Disentangled Representation Learning with Analogical Relations