Zobrazeno 1 - 10
of 937
pro vyhledávání: '"Roth, Kevin A"'
Publikováno v:
NeurIPS 2021
Recent works on Bayesian neural networks (BNNs) have highlighted the need to better understand the implications of using Gaussian priors in combination with the compositional structure of the network architecture. Similar in spirit to the kind of ana
Externí odkaz:
http://arxiv.org/abs/2106.06615
Publikováno v:
NeurIPS 2021
The "cold posterior effect" (CPE) in Bayesian deep learning describes the uncomforting observation that the predictive performance of Bayesian neural networks can be significantly improved if the Bayes posterior is artificially sharpened using a temp
Externí odkaz:
http://arxiv.org/abs/2106.06596
Autor:
Roth, Kevin
The existence of adversarial examples poses a real danger when deep neural networks are deployed in the real world. The go-to strategy to quantify this vulnerability is to evaluate the model against specific attack algorithms. This approach is howeve
Externí odkaz:
http://arxiv.org/abs/2106.03099
Autor:
Swiatkowski, Jakub, Roth, Kevin, Veeling, Bastiaan S., Tran, Linh, Dillon, Joshua V., Snoek, Jasper, Mandt, Stephan, Salimans, Tim, Jenatton, Rodolphe, Nowozin, Sebastian
Variational Bayesian Inference is a popular methodology for approximating posterior distributions over Bayesian neural network weights. Recent work developing this class of methods has explored ever richer parameterizations of the approximate posteri
Externí odkaz:
http://arxiv.org/abs/2002.02655
Autor:
Wenzel, Florian, Roth, Kevin, Veeling, Bastiaan S., Świątkowski, Jakub, Tran, Linh, Mandt, Stephan, Snoek, Jasper, Salimans, Tim, Jenatton, Rodolphe, Nowozin, Sebastian
During the past five years the Bayesian deep learning community has developed increasingly accurate and efficient approximate inference procedures that allow for Bayesian inference in deep neural networks. However, despite this algorithmic progress a
Externí odkaz:
http://arxiv.org/abs/2002.02405
Autor:
Tran, Linh, Veeling, Bastiaan S., Roth, Kevin, Swiatkowski, Jakub, Dillon, Joshua V., Snoek, Jasper, Mandt, Stephan, Salimans, Tim, Nowozin, Sebastian, Jenatton, Rodolphe
Ensembles of models have been empirically shown to improve predictive performance and to yield robust measures of uncertainty. However, they are expensive in computation and memory. Therefore, recent research has focused on distilling ensembles into
Externí odkaz:
http://arxiv.org/abs/2001.04694
We establish a theoretical link between adversarial training and operator norm regularization for deep neural networks. Specifically, we prove that $\ell_p$-norm constrained projected gradient ascent based adversarial training with an $\ell_q$-norm l
Externí odkaz:
http://arxiv.org/abs/1906.01527
We investigate conditions under which test statistics exist that can reliably detect examples, which have been adversarially manipulated in a white-box attack. These statistics can be easily computed and calibrated by randomly corrupting inputs. They
Externí odkaz:
http://arxiv.org/abs/1902.04818
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
We propose a novel data-dependent structured gradient regularizer to increase the robustness of neural networks vis-a-vis adversarial perturbations. Our regularizer can be derived as a controlled approximation from first principles, leveraging the fu
Externí odkaz:
http://arxiv.org/abs/1805.08736