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

Machine learning models are typically made available to potential clientusers via inference APIs. Model extraction attacks occur when a maliciousclient uses information gleaned from queries to the inference API of a victimmodel $F_V$ to build a surrogate model $F_A$ with comparable functionality.Recent research has shown successful model extraction of image classification,and natural language processing models. In this paper, we show the first modelextraction attack against real-world generative adversarial network (GAN) imagetranslation models. We present a framework for conducting such attacks, andshow that an adversary can successfully extract functional surrogate models byquerying $F_V$ using data from the same domain as the training data for $F_V$.The adversary need not know $F_V$'s architecture or any other information aboutit beyond its intended task. We evaluate the effectiveness of our attacks usingthree different instances of two popular categories of image translation: (1)Selfie-to-Anime and (2) Monet-to-Photo (image style transfer), and (3)Super-Resolution (super resolution). Using standard performance metrics forGANs, we show that our attacks are effective. Furthermore, we conducted a largescale (125 participants) user study on Selfie-to-Anime and Monet-to-Photo toshow that human perception of the images produced by $F_V$ and $F_A$ can beconsidered equivalent, within an equivalence bound of Cohen's d = 0.3. Finally,we show that existing defenses against model extraction attacks (watermarking,adversarial examples, poisoning) do not extend to image translation models.

Good Artists Copy, Great Artists Steal: Model Extraction Attacks Against Image Translation Models