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

The interplay between quantum physics and machine learning gives rise to theemergent frontier of quantum machine learning, where advanced quantum learningmodels may outperform their classical counterparts in solving certainchallenging problems. However, quantum learning systems are vulnerable toadversarial attacks: adding tiny carefully-crafted perturbations on legitimateinput samples can cause misclassifications. To address this issue, we propose ageneral scheme to protect quantum learning systems from adversarial attacks byrandomly encoding the legitimate data samples through unitary or quantum errorcorrection encoders. In particular, we rigorously prove that both global andlocal random unitary encoders lead to exponentially vanishing gradients (i.e.barren plateaus) for any variational quantum circuits that aim to addadversarial perturbations, independent of the input data and the innerstructures of adversarial circuits and quantum classifiers. In addition, weprove a rigorous bound on the vulnerability of quantum classifiers under localunitary adversarial attacks. We show that random black-box quantum errorcorrection encoders can protect quantum classifiers against local adversarialnoises and their robustness increases as we concatenate error correction codes.To quantify the robustness enhancement, we adapt quantum differential privacyas a measure of the prediction stability for quantum classifiers. Our resultsestablish versatile defense strategies for quantum classifiers againstadversarial perturbations, which provide valuable guidance to enhance thereliability and security for both near-term and future quantum learningtechnologies.

Enhancing Quantum Adversarial Robustness by Randomized Encodings