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

Previous studies have revealed that artificial intelligence (AI) systems arevulnerable to adversarial attacks. Among them, model extraction attacks foolthe target model by generating adversarial examples on a substitute model. Thecore of such an attack is training a substitute model as similar to the targetmodel as possible, where the simulation process can be categorized in adata-dependent and data-free manner. Compared with the data-dependent method,the data-free one has been proven to be more practical in the real world sinceit trains the substitute model with synthesized data. However, the distributionof these fake data lacks diversity and cannot detect the decision boundary ofthe target model well, resulting in the dissatisfactory simulation effect.Besides, these data-free techniques need a vast number of queries to train thesubstitute model, increasing the time and computing consumption and the risk ofexposure. To solve the aforementioned problems, in this paper, we propose anovel data-free model extraction method named SCME (Self-Contrastive ModelExtraction), which considers both the inter- and intra-class diversity insynthesizing fake data. In addition, SCME introduces the Mixup operation toaugment the fake data, which can explore the target model's decision boundaryeffectively and improve the simulating capacity. Extensive experiments showthat the proposed method can yield diversified fake data. Moreover, our methodhas shown superiority in many different attack settings under the query-limitedscenario, especially for untargeted attacks, the SCME outperforms SOTA methodsby 11.43\% on average for five baseline datasets.

SCME: A Self-Contrastive Method for Data-free and Query-Limited Model Extraction Attack