Zobrazeno 1 - 10
of 29
pro vyhledávání: '"Qin, Chongli"'
Autor:
Jiralerspong, Marco, Bose, Avishek Joey, Gemp, Ian, Qin, Chongli, Bachrach, Yoram, Gidel, Gauthier
The past few years have seen impressive progress in the development of deep generative models capable of producing high-dimensional, complex, and photo-realistic data. However, current methods for evaluating such models remain incomplete: standard li
Externí odkaz:
http://arxiv.org/abs/2302.04440
The recipe behind the success of deep learning has been the combination of neural networks and gradient-based optimization. Understanding the behavior of gradient descent however, and particularly its instability, has lagged behind its empirical succ
Externí odkaz:
http://arxiv.org/abs/2302.01952
Autor:
Qin, Chongli, Wu, Yan, Springenberg, Jost Tobias, Brock, Andrew, Donahue, Jeff, Lillicrap, Timothy P., Kohli, Pushmeet
The instability of Generative Adversarial Network (GAN) training has frequently been attributed to gradient descent. Consequently, recent methods have aimed to tailor the models and training procedures to stabilise the discrete updates. In contrast,
Externí odkaz:
http://arxiv.org/abs/2010.15040
Adversarial training and its variants have become de facto standards for learning robust deep neural networks. In this paper, we explore the landscape around adversarial training in a bid to uncover its limits. We systematically study the effect of d
Externí odkaz:
http://arxiv.org/abs/2010.03593
Autor:
Gowal, Sven, Qin, Chongli, Huang, Po-Sen, Cemgil, Taylan, Dvijotham, Krishnamurthy, Mann, Timothy, Kohli, Pushmeet
Recent research has made the surprising finding that state-of-the-art deep learning models sometimes fail to generalize to small variations of the input. Adversarial training has been shown to be an effective approach to overcome this problem. Howeve
Externí odkaz:
http://arxiv.org/abs/1912.03192
Adversarial testing methods based on Projected Gradient Descent (PGD) are widely used for searching norm-bounded perturbations that cause the inputs of neural networks to be misclassified. This paper takes a deeper look at these methods and explains
Externí odkaz:
http://arxiv.org/abs/1910.09338
Autor:
Qin, Chongli, Martens, James, Gowal, Sven, Krishnan, Dilip, Dvijotham, Krishnamurthy, Fawzi, Alhussein, De, Soham, Stanforth, Robert, Kohli, Pushmeet
Adversarial training is an effective methodology for training deep neural networks that are robust against adversarial, norm-bounded perturbations. However, the computational cost of adversarial training grows prohibitively as the size of the model a
Externí odkaz:
http://arxiv.org/abs/1907.02610
Autor:
Qin, Chongli, Krishnamurthy, Dvijotham, O'Donoghue, Brendan, Bunel, Rudy, Stanforth, Robert, Gowal, Sven, Uesato, Jonathan, Swirszcz, Grzegorz, Kohli, Pushmeet
Prior work on neural network verification has focused on specifications that are linear functions of the output of the network, e.g., invariance of the classifier output under adversarial perturbations of the input. In this paper, we extend verificat
Externí odkaz:
http://arxiv.org/abs/1902.09592
Autor:
Gowal, Sven, Dvijotham, Krishnamurthy, Stanforth, Robert, Bunel, Rudy, Qin, Chongli, Uesato, Jonathan, Arandjelovic, Relja, Mann, Timothy, Kohli, Pushmeet
Recent work has shown that it is possible to train deep neural networks that are provably robust to norm-bounded adversarial perturbations. Most of these methods are based on minimizing an upper bound on the worst-case loss over all possible adversar
Externí odkaz:
http://arxiv.org/abs/1810.12715
Autor:
Qin, Chongli, Colwell, Lucy J.
Publikováno v:
Proceedings of the National Academy of Sciences of the United States of America, 2018 Jan . 115(4), 690-695.
Externí odkaz:
https://www.jstor.org/stable/26506455