Guy Amit (Ben-Gurion University), Moshe Levy (Ben-Gurion University), Yisroel Mirsky (Ben-Gurion University)

Deep neural networks are normally executed in the forward direction. However, in this work, we identify a vulnerability that enables models to be trained in both directions and on different tasks. Adversaries can exploit this capability to hide rogue models within seemingly legitimate models. In addition, in this work we show that neural networks can be taught to systematically memorize and retrieve specific samples from datasets. Together, these findings expose a novel method in which adversaries can exfiltrate datasets from protected learning environments under the guise of legitimate models.

We focus on the data exfiltration attack and show that modern architectures can be used to secretly exfiltrate tens of thousands of samples with high fidelity, high enough to compromise data privacy and even train new models. Moreover, to mitigate this threat we propose a novel approach for detecting infected models.

View More Papers

Analysis of the Effect of the Difference between Japanese...

Rei Yamagishi, Shinya Sasa, and Shota Fujii (Hitachi, Ltd.)

Read More

A Security and Usability Analysis of Local Attacks Against...

Tarun Kumar Yadav (Brigham Young University), Kent Seamons (Brigham Young University)

Read More

Exploring Phishing Threats through QR Codes in Naturalistic Settings

Filipo Sharevski (DePaul University), Mattia Mossano, Maxime Fabian Veit, Gunther Schiefer, Melanie Volkamer (Karlsruhe Institute of Technology)

Read More

ActiveDaemon: Unconscious DNN Dormancy and Waking Up via User-specific...

Ge Ren (Shanghai Jiao Tong University), Gaolei Li (Shanghai Jiao Tong University), Shenghong Li (Shanghai Jiao Tong University), Libo Chen (Shanghai Jiao Tong University), Kui Ren (Zhejiang University)

Read More