Výsledky vyhledávání

Automated Lane Centering (ALC) systems are convenient and widely deployed today, but also highly security and safety critical. In this work, we are the first to systematically study the security of state-of-the-art deep learning based ALC systems in

Externí odkaz: http://arxiv.org/abs/2009.06701

Zobrazit plný text záznamu

Report

Towards Practical Lottery Ticket Hypothesis for Adversarial Training

Autor: Li, Bai, Wang, Shiqi, Jia, Yunhan, Lu, Yantao, Zhong, Zhenyu, Carin, Lawrence, Jana, Suman

Recent research has proposed the lottery ticket hypothesis, suggesting that for a deep neural network, there exist trainable sub-networks performing equally or better than the original model with commensurate training steps. While this discovery is i

Externí odkaz: http://arxiv.org/abs/2003.05733

Zobrazit plný text záznamu

Report

Security of Deep Learning based Lane Keeping System under Physical-World Adversarial Attack

Autor: Sato, Takami, Shen, Junjie, Wang, Ningfei, Jia, Yunhan Jack, Lin, Xue, Chen, Qi Alfred

Lane-Keeping Assistance System (LKAS) is convenient and widely available today, but also extremely security and safety critical. In this work, we design and implement the first systematic approach to attack real-world DNN-based LKASes. We identify di

Externí odkaz: http://arxiv.org/abs/2003.01782

Zobrazit plný text záznamu

Akademický článek

Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.

Report

Enhancing Cross-task Black-Box Transferability of Adversarial Examples with Dispersion Reduction

Autor: Lu, Yantao, Jia, Yunhan, Wang, Jianyu, Li, Bai, Chai, Weiheng, Carin, Lawrence, Velipasalar, Senem

Neural networks are known to be vulnerable to carefully crafted adversarial examples, and these malicious samples often transfer, i.e., they remain adversarial even against other models. Although great efforts have been delved into the transferabilit

Externí odkaz: http://arxiv.org/abs/1911.11616

Zobrazit plný text záznamu

Report

Explainable Machine Learning in Deployment

Autor: Bhatt, Umang, Xiang, Alice, Sharma, Shubham, Weller, Adrian, Taly, Ankur, Jia, Yunhan, Ghosh, Joydeep, Puri, Ruchir, Moura, José M. F., Eckersley, Peter

Explainable machine learning offers the potential to provide stakeholders with insights into model behavior by using various methods such as feature importance scores, counterfactual explanations, or influential training data. Yet there is little und

Externí odkaz: http://arxiv.org/abs/1909.06342

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání