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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

One major factor impeding more widespread adoption of deep neural networks(DNNs) is their lack of robustness, which is essential for safety-criticalapplications such as autonomous driving. This has motivated much recent work onadversarial attacks for DNNs, which mostly focus on pixel-level perturbationsvoid of semantic meaning. In contrast, we present a general framework foradversarial attacks on trained agents, which covers semantic perturbations tothe environment of the agent performing the task as well as pixel-levelattacks. To do this, we re-frame the adversarial attack problem as learning adistribution of parameters that always fools the agent. In the semantic case,our proposed adversary (denoted as BBGAN) is trained to sample parameters thatdescribe the environment with which the black-box agent interacts, such thatthe agent performs its dedicated task poorly in this environment. We applyBBGAN on three different tasks, primarily targeting aspects of autonomousnavigation: object detection, self-driving, and autonomous UAV racing. On thesetasks, BBGAN can generate failure cases that consistently fool a trained agent.

SADA: Semantic Adversarial Diagnostic Attacks for Autonomous Applications