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

Convolutional neural networks encode images through a sequence ofconvolutions, normalizations and non-linearities as well as downsamplingoperations into potentially strong semantic embeddings. Yet, previous workshowed that even slight mistakes during sampling, leading to aliasing, can bedirectly attributed to the networks' lack in robustness. To address such issuesand facilitate simpler and faster adversarial training, [12] recently proposedFLC pooling, a method for provably alias-free downsampling - in theory. In thiswork, we conduct a further analysis through the lens of signal processing andfind that such current pooling methods, which address aliasing in the frequencydomain, are still prone to spectral leakage artifacts. Hence, we proposealiasing and spectral artifact-free pooling, short ASAP. While only introducinga few modifications to FLC pooling, networks using ASAP as downsampling methodexhibit higher native robustness against common corruptions, a property thatFLC pooling was missing. ASAP also increases native robustness againstadversarial attacks on high and low resolution data while maintaining similarclean accuracy or even outperforming the baseline.

Fix your downsampling ASAP! Be natively more robust via Aliasing and Spectral Artifact free Pooling