A
Amin Ghiasi
Researcher at University of Maryland, College Park
Publications - 16
Citations - 467
Amin Ghiasi is an academic researcher from University of Maryland, College Park. The author has contributed to research in topics: Computer science & Transfer of learning. The author has an hindex of 7, co-authored 13 publications receiving 340 citations.
Papers
More filters
Posted Content
Adversarial Training for Free
Ali Shafahi,Mahyar Najibi,Amin Ghiasi,Zheng Xu,John P. Dickerson,Christoph Studer,Larry S. Davis,Gavin Taylor,Tom Goldstein +8 more
TL;DR: This work presents an algorithm that eliminates the overhead cost of generating adversarial examples by recycling the gradient information computed when updating model parameters, and achieves comparable robustness to PGD adversarial training on the CIFAR-10 and CIFar-100 datasets at negligible additional cost compared to natural training.
Posted Content
Adversarially robust transfer learning
Ali Shafahi,Parsa Saadatpanah,Chen Zhu,Amin Ghiasi,Christoph Studer,David W. Jacobs,Tom Goldstein +6 more
TL;DR: This work considers robust transfer learning, in which not only performance but also robustness from a source model to a target domain is transferred, and can improve the generalization of adversarially trained models, while maintaining their robustness.
Posted Content
Strong Data Augmentation Sanitizes Poisoning and Backdoor Attacks Without an Accuracy Tradeoff
Eitan Borgnia,Valeriia Cherepanova,Liam Fowl,Amin Ghiasi,Jonas Geiping,Micah Goldblum,Tom Goldstein,Arjun Gupta +7 more
TL;DR: It is found that strong data augmentations, such as mixup and CutMix, can significantly diminish the threat of poisoning and backdoor attacks without trading off performance.
Proceedings Article
Adversarially robust transfer learning
Ali Shafahi,Parsa Saadatpanah,Chen Zhu,Amin Ghiasi,Christoph Studer,David W. Jacobs,Tom Goldstein +6 more
TL;DR: In this article, robust transfer learning is used to produce a model that is not only accurate but also adversarially robust, where data scarcity and computational limitations become even more cumbersome.
Posted Content
Label Smoothing and Logit Squeezing: A Replacement for Adversarial Training?
TL;DR: This study explores the mechanisms by which adversarial training improves classifier robustness, and shows that these mechanisms can be effectively mimicked using simple regularization methods, including label smoothing and logit squeezing.