An Effective Neural Network Model Protection Method Against Model Stealing Attacks for Image Classification Applications

研究成果: Conference contribution

摘要

The growing adoption of Deep Learning (DL) models at the network edge necessitates robust security measures. Image preprocessing, a common technique for securing DL models, often involves shuffling the pixels of training images before feeding them to the model. In this work, we found that the shuffle-then-flip method offers superior resilience compared to the shuffle-only method. We also investigate the effectiveness of the shuffle-then-flip preprocessing technique in enhancing the security of models against attacks.

原文English
主出版物標題Proceedings - International SoC Design Conference 2024, ISOCC 2024
發行者Institute of Electrical and Electronics Engineers Inc.
頁面125-126
頁數2
ISBN(電子)9798350377088
DOIs
出版狀態Published - 2024
事件21st International System-on-Chip Design Conference, ISOCC 2024 - Sapporo, Japan
持續時間: 2024 8月 192024 8月 22

出版系列

名字Proceedings - International SoC Design Conference 2024, ISOCC 2024

Conference

Conference21st International System-on-Chip Design Conference, ISOCC 2024
國家/地區Japan
城市Sapporo
期間24-08-1924-08-22

All Science Journal Classification (ASJC) codes

  • 人工智慧
  • 硬體和架構
  • 能源工程與電力技術
  • 電氣與電子工程
  • 電子、光磁材料

指紋

深入研究「An Effective Neural Network Model Protection Method Against Model Stealing Attacks for Image Classification Applications」主題。共同形成了獨特的指紋。

引用此