WaNet - Imperceptible Warping-based Backdoor Attack
3. května 2021
Řečníci
Anh Nguyen
Řečník · 0 sledujících
O prezentaci
With the thriving of deep learning and the widespread practice of using pre-trained networks, backdoor attacks have become an increasing security threat drawing many research interests in recent years. A third-party model can be poisoned in training to work well in normal conditions but behave maliciously when a trigger pattern appears. However, the existing backdoor attacks are all built on noise perturbation triggers, making them noticeable to humans. In this paper, we instead propose using warping-based triggers. The proposed backdoor outperforms the previous methods in a human inspection test by a wide margin, proving its stealthiness. To make such models undetectable by machine defenders, we propose a novel training mode, called the ``noise mode. The trained networks successfully attack and bypass the state of the art defense methods on standard classification datasets, including MNIST, CIFAR-10, GTSRB, and CelebA. Behavior analyses show that our backdoors are transparent to network inspection, further proving this novel attack mechanism's efficiency.With the thriving of deep learning and the widespread practice of using pre-trained networks, backdoor attacks have become an increasing security threat drawing many research interests in recent years. A third-party model can be poisoned in training to work well in normal conditions but behave maliciously when a trigger pattern appears. However, the existing backdoor attacks are all built on noise perturbation triggers, making them noticeable to humans. In this paper, we instead propose using wa…
Kategorie
Umělá inteligence a data science
Kategorie · 10,8k prezentací
O organizátorovi (ICLR 2021)
The International Conference on Learning Representations (ICLR) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence called representation learning, but generally referred to as deep learning. ICLR is globally renowned for presenting and publishing cutting-edge research on all aspects of deep learning used in the fields of artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, text understanding, gaming, and robotics.
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
Unlearnable Examples: Making Personal Data Unexploitable
Zhlédnout později
Hanxun Huang, …
Oral Session 5
Zhlédnout později
Decentralized Attribution of Generative Models
Zhlédnout později
Ask Your Humans: Using Human Instructions to Improve Generalization in Reinforcement Learning
Zhlédnout později
Valerie Chen, …
Announcement of the different award winners and closing remarks
Zhlédnout později
Measuring Adversarial Robustness using a Voronoi-Epsilon Adversary
Zhlédnout později
Hyeongji Kim, …