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

Deep learning algorithms have been shown to perform extremely well on manyclassical machine learning problems. However, recent studies have shown thatdeep learning, like other machine learning techniques, is vulnerable toadversarial samples: inputs crafted to force a deep neural network (DNN) toprovide adversary-selected outputs. Such attacks can seriously undermine thesecurity of the system supported by the DNN, sometimes with devastatingconsequences. For example, autonomous vehicles can be crashed, illicit orillegal content can bypass content filters, or biometric authentication systemscan be manipulated to allow improper access. In this work, we introduce adefensive mechanism called defensive distillation to reduce the effectivenessof adversarial samples on DNNs. We analytically investigate thegeneralizability and robustness properties granted by the use of defensivedistillation when training DNNs. We also empirically study the effectiveness ofour defense mechanisms on two DNNs placed in adversarial settings. The studyshows that defensive distillation can reduce effectiveness of sample creationfrom 95% to less than 0.5% on a studied DNN. Such dramatic gains can beexplained by the fact that distillation leads gradients used in adversarialsample creation to be reduced by a factor of 10^30. We also find thatdistillation increases the average minimum number of features that need to bemodified to create adversarial samples by about 800% on one of the DNNs wetested.

Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks