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

Variational autoencoders (VAEs) are popular models for representationlearning but their encoders are susceptible to overfitting (Cremer et al.,2018) because they are trained on a finite training set instead of the true(continuous) data distribution $p_{\mathrm{data}}(\mathbf{x})$. Diffusionmodels, on the other hand, avoid this issue by keeping the encoder fixed. Thismakes their representations less interpretable, but it simplifies training,enabling accurate and continuous approximations of$p_{\mathrm{data}}(\mathbf{x})$. In this paper, we show that overfittingencoders in VAEs can be effectively mitigated by training on samples from apre-trained diffusion model. These results are somewhat unexpected as recentfindings (Alemohammad et al., 2023; Shumailov et al., 2023) observe a decay ingenerative performance when models are trained on data generated by anothergenerative model. We analyze generalization performance, amortization gap, androbustness of VAEs trained with our proposed method on three different datasets. We find improvements in all metrics compared to both normal training andconventional data augmentation methods, and we show that a modest amount ofsamples from the diffusion model suffices to obtain these gains.

Upgrading VAE Training With Unlimited Data Plans Provided by Diffusion Models