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

Recent research has made the surprising finding that state-of-the-art deeplearning models sometimes fail to generalize to small variations of the input.Adversarial training has been shown to be an effective approach to overcomethis problem. However, its application has been limited to enforcing invarianceto analytically defined transformations like $\ell_p$-norm boundedperturbations. Such perturbations do not necessarily cover plausible real-worldvariations that preserve the semantics of the input (such as a change inlighting conditions). In this paper, we propose a novel approach to express andformalize robustness to these kinds of real-world transformations of the input.The two key ideas underlying our formulation are (1) leveraging disentangledrepresentations of the input to define different factors of variations, and (2)generating new input images by adversarially composing the representations ofdifferent images. We use a StyleGAN model to demonstrate the efficacy of thisframework. Specifically, we leverage the disentangled latent representationscomputed by a StyleGAN model to generate perturbations of an image that aresimilar to real-world variations (like adding make-up, or changing theskin-tone of a person) and train models to be invariant to these perturbations.Extensive experiments show that our method improves generalization and reducesthe effect of spurious correlations (reducing the error rate of a "smile"detector by 21% for example).

Achieving Robustness in the Wild via Adversarial Mixing with Disentangled Representations