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

Variational AutoEncoders (VAEs) provide a means to generate representationallatent embeddings. Previous research has highlighted the benefits of achievingrepresentations that are disentangled, particularly for downstream tasks.However, there is some debate about how to encourage disentanglement with VAEsand evidence indicates that existing implementations of VAEs do not achievedisentanglement consistently. The evaluation of how well a VAE's latent spacehas been disentangled is often evaluated against our subjective expectations ofwhich attributes should be disentangled for a given problem. Therefore, bydefinition, we already have domain knowledge of what should be achieved and yetwe use unsupervised approaches to achieve it. We propose a weakly-supervisedapproach that incorporates any available domain knowledge into the trainingprocess to form a Gated-VAE. The process involves partitioning therepresentational embedding and gating backpropagation. All partitions areutilised on the forward pass but gradients are backpropagated through differentpartitions according to selected image/target pairings. The approach can beused to modify existing VAE models such as beta-VAE, InfoVAE and DIP-VAE-II.Experiments demonstrate that using gated backpropagation, latent factors arerepresented in their intended partition. The approach is applied to images offaces for the purpose of disentangling head-pose from facial expression.Quantitative metrics show that using Gated-VAE improves averagedisentanglement, completeness and informativeness, as compared with un-gatedimplementations. Qualitative assessment of latent traversals demonstrate itsdisentanglement of head-pose from expression, even when only weak/noisysupervision is available.

Gated Variational AutoEncoders: Incorporating Weak Supervision to Encourage Disentanglement