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

Evasion attacks against machine learning models often succeed via iterativeprobing of a fixed target model, whereby an attack that succeeds once willsucceed repeatedly. One promising approach to counter this threat is making amodel a moving target against adversarial inputs. To this end, we introduceMorphence-2.0, a scalable moving target defense (MTD) powered byout-of-distribution (OOD) detection to defend against adversarial examples. Byregularly moving the decision function of a model, Morphence-2.0 makes itsignificantly challenging for repeated or correlated attacks to succeed.Morphence-2.0 deploys a pool of models generated from a base model in a mannerthat introduces sufficient randomness when it responds to prediction queries.Via OOD detection, Morphence-2.0 is equipped with a scheduling approach thatassigns adversarial examples to robust decision functions and benign samples toan undefended accurate models. To ensure repeated or correlated attacks fail,the deployed pool of models automatically expires after a query budget isreached and the model pool is seamlessly replaced by a new model pool generatedin advance. We evaluate Morphence-2.0 on two benchmark image classificationdatasets (MNIST and CIFAR10) against 4 reference attacks (3 white-box and 1black-box). Morphence-2.0 consistently outperforms prior defenses whilepreserving accuracy on clean data and reducing attack transferability. We alsoshow that, when powered by OOD detection, Morphence-2.0 is able to preciselymake an input-based movement of the model's decision function that leads tohigher prediction accuracy on both adversarial and benign queries.

Morphence-2.0: Evasion-Resilient Moving Target Defense Powered by Out-of-Distribution Detection