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

Existing deep neural networks (DNNs) that achieve state-of-the-art (SOTA)performance on both clean and adversarially-perturbed images rely on eitheractivation or weight conditioned convolution operations. However, suchconditional learning costs additional multiply-accumulate (MAC) or additionoperations, increasing inference memory and compute costs. To that end, wepresent a sparse mixture once for all adversarial training (SMART), that allowsa model to train once and then in-situ trade-off between accuracy androbustness, that too at a reduced compute and parameter overhead. Inparticular, SMART develops two expert paths, for clean and adversarial images,respectively, that are then conditionally trained via respective dedicated setsof binary sparsity masks. Extensive evaluations on multiple imageclassification datasets across different models show SMART to have up to 2.72xfewer non-zero parameters costing proportional reduction in compute overhead,while yielding SOTA accuracy-robustness trade-off. Additionally, we presentinsightful observations in designing sparse masks to successfully condition onboth clean and perturbed images.

Sparse Mixture Once-for-all Adversarial Training for Efficient In-Situ Trade-Off Between Accuracy and Robustness of DNNs