The Experts below are selected from a list of 72 Experts worldwide ranked by ideXlab platform
Stanley Osher - One of the best experts on this subject based on the ideXlab platform.
-
adversarial defense via the data dependent Activation total variation minimization and adversarial training
Inverse Problems and Imaging, 2021Co-Authors: Bao Wang, Wei Zhu, Alex Tong Lin, Penghang Yin, Andrea L Bertozzi, Stanley OsherAbstract:We improve the robustness of Deep Neural Net (DNN) to adversarial attacks by using an interpolating function as the Output Activation. This data-dependent Activation remarkably improves both the generalization and robustness of DNN. In the CIFAR10 benchmark, we raise the robust accuracy of the adversarially trained ResNet20 from \begin{document}$ \sim 46\% $\end{document} to \begin{document}$ \sim 69\% $\end{document} under the state-of-the-art Iterative Fast Gradient Sign Method (IFGSM) based adversarial attack. When we combine this data-dependent Activation with total variation minimization on adversarial images and training data augmentation, we achieve an improvement in robust accuracy by 38.9 \begin{document}$ \% $\end{document} for ResNet56 under the strongest IFGSM attack. Furthermore, We provide an intuitive explanation of our defense by analyzing the geometry of the feature space.
-
adversarial defense via data dependent Activation function and total variation minimization
arXiv: Learning, 2018Co-Authors: Bao Wang, Wei Zhu, Zuoqiang Shi, Alex Tong Lin, Penghang Yin, Andrea L Bertozzi, Stanley OsherAbstract:Author(s): Wang, Bao; Lin, Alex T; Zhu, Wei; Yin, Penghang; Bertozzi, Andrea L; Osher, Stanley J | Abstract: We improve the robustness of Deep Neural Net (DNN) to adversarial attacks by using an interpolating function as the Output Activation. This data-dependent Activation remarkably improves both the generalization and robustness of DNN. In the CIFAR10 benchmark, we raise the robust accuracy of the adversarially trained ResNet20 from $\sim 46\%$ to $\sim 69\%$ under the state-of-the-art Iterative Fast Gradient Sign Method (IFGSM) based adversarial attack. When we combine this data-dependent Activation with total variation minimization on adversarial images and training data augmentation, we achieve an improvement in robust accuracy by 38.9$\%$ for ResNet56 under the strongest IFGSM attack. Furthermore, We provide an intuitive explanation of our defense by analyzing the geometry of the feature space.
-
Deep Neural Nets with Interpolating Function as Output Activation
arXiv: Learning, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and code will be made publicly available.
-
deep neural nets with interpolating function as Output Activation
Neural Information Processing Systems, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and the code is available at https://github.com/ BaoWangMath/DNN-DataDependentActivation.
Bao Wang - One of the best experts on this subject based on the ideXlab platform.
-
adversarial defense via the data dependent Activation total variation minimization and adversarial training
Inverse Problems and Imaging, 2021Co-Authors: Bao Wang, Wei Zhu, Alex Tong Lin, Penghang Yin, Andrea L Bertozzi, Stanley OsherAbstract:We improve the robustness of Deep Neural Net (DNN) to adversarial attacks by using an interpolating function as the Output Activation. This data-dependent Activation remarkably improves both the generalization and robustness of DNN. In the CIFAR10 benchmark, we raise the robust accuracy of the adversarially trained ResNet20 from \begin{document}$ \sim 46\% $\end{document} to \begin{document}$ \sim 69\% $\end{document} under the state-of-the-art Iterative Fast Gradient Sign Method (IFGSM) based adversarial attack. When we combine this data-dependent Activation with total variation minimization on adversarial images and training data augmentation, we achieve an improvement in robust accuracy by 38.9 \begin{document}$ \% $\end{document} for ResNet56 under the strongest IFGSM attack. Furthermore, We provide an intuitive explanation of our defense by analyzing the geometry of the feature space.
-
adversarial defense via data dependent Activation function and total variation minimization
arXiv: Learning, 2018Co-Authors: Bao Wang, Wei Zhu, Zuoqiang Shi, Alex Tong Lin, Penghang Yin, Andrea L Bertozzi, Stanley OsherAbstract:Author(s): Wang, Bao; Lin, Alex T; Zhu, Wei; Yin, Penghang; Bertozzi, Andrea L; Osher, Stanley J | Abstract: We improve the robustness of Deep Neural Net (DNN) to adversarial attacks by using an interpolating function as the Output Activation. This data-dependent Activation remarkably improves both the generalization and robustness of DNN. In the CIFAR10 benchmark, we raise the robust accuracy of the adversarially trained ResNet20 from $\sim 46\%$ to $\sim 69\%$ under the state-of-the-art Iterative Fast Gradient Sign Method (IFGSM) based adversarial attack. When we combine this data-dependent Activation with total variation minimization on adversarial images and training data augmentation, we achieve an improvement in robust accuracy by 38.9$\%$ for ResNet56 under the strongest IFGSM attack. Furthermore, We provide an intuitive explanation of our defense by analyzing the geometry of the feature space.
-
Deep Neural Nets with Interpolating Function as Output Activation
arXiv: Learning, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and code will be made publicly available.
-
deep neural nets with interpolating function as Output Activation
Neural Information Processing Systems, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and the code is available at https://github.com/ BaoWangMath/DNN-DataDependentActivation.
Wei Zhu - One of the best experts on this subject based on the ideXlab platform.
-
adversarial defense via the data dependent Activation total variation minimization and adversarial training
Inverse Problems and Imaging, 2021Co-Authors: Bao Wang, Wei Zhu, Alex Tong Lin, Penghang Yin, Andrea L Bertozzi, Stanley OsherAbstract:We improve the robustness of Deep Neural Net (DNN) to adversarial attacks by using an interpolating function as the Output Activation. This data-dependent Activation remarkably improves both the generalization and robustness of DNN. In the CIFAR10 benchmark, we raise the robust accuracy of the adversarially trained ResNet20 from \begin{document}$ \sim 46\% $\end{document} to \begin{document}$ \sim 69\% $\end{document} under the state-of-the-art Iterative Fast Gradient Sign Method (IFGSM) based adversarial attack. When we combine this data-dependent Activation with total variation minimization on adversarial images and training data augmentation, we achieve an improvement in robust accuracy by 38.9 \begin{document}$ \% $\end{document} for ResNet56 under the strongest IFGSM attack. Furthermore, We provide an intuitive explanation of our defense by analyzing the geometry of the feature space.
-
adversarial defense via data dependent Activation function and total variation minimization
arXiv: Learning, 2018Co-Authors: Bao Wang, Wei Zhu, Zuoqiang Shi, Alex Tong Lin, Penghang Yin, Andrea L Bertozzi, Stanley OsherAbstract:Author(s): Wang, Bao; Lin, Alex T; Zhu, Wei; Yin, Penghang; Bertozzi, Andrea L; Osher, Stanley J | Abstract: We improve the robustness of Deep Neural Net (DNN) to adversarial attacks by using an interpolating function as the Output Activation. This data-dependent Activation remarkably improves both the generalization and robustness of DNN. In the CIFAR10 benchmark, we raise the robust accuracy of the adversarially trained ResNet20 from $\sim 46\%$ to $\sim 69\%$ under the state-of-the-art Iterative Fast Gradient Sign Method (IFGSM) based adversarial attack. When we combine this data-dependent Activation with total variation minimization on adversarial images and training data augmentation, we achieve an improvement in robust accuracy by 38.9$\%$ for ResNet56 under the strongest IFGSM attack. Furthermore, We provide an intuitive explanation of our defense by analyzing the geometry of the feature space.
-
Deep Neural Nets with Interpolating Function as Output Activation
arXiv: Learning, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and code will be made publicly available.
-
deep neural nets with interpolating function as Output Activation
Neural Information Processing Systems, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and the code is available at https://github.com/ BaoWangMath/DNN-DataDependentActivation.
Zuoqiang Shi - One of the best experts on this subject based on the ideXlab platform.
-
adversarial defense via data dependent Activation function and total variation minimization
arXiv: Learning, 2018Co-Authors: Bao Wang, Wei Zhu, Zuoqiang Shi, Alex Tong Lin, Penghang Yin, Andrea L Bertozzi, Stanley OsherAbstract:Author(s): Wang, Bao; Lin, Alex T; Zhu, Wei; Yin, Penghang; Bertozzi, Andrea L; Osher, Stanley J | Abstract: We improve the robustness of Deep Neural Net (DNN) to adversarial attacks by using an interpolating function as the Output Activation. This data-dependent Activation remarkably improves both the generalization and robustness of DNN. In the CIFAR10 benchmark, we raise the robust accuracy of the adversarially trained ResNet20 from $\sim 46\%$ to $\sim 69\%$ under the state-of-the-art Iterative Fast Gradient Sign Method (IFGSM) based adversarial attack. When we combine this data-dependent Activation with total variation minimization on adversarial images and training data augmentation, we achieve an improvement in robust accuracy by 38.9$\%$ for ResNet56 under the strongest IFGSM attack. Furthermore, We provide an intuitive explanation of our defense by analyzing the geometry of the feature space.
-
Deep Neural Nets with Interpolating Function as Output Activation
arXiv: Learning, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and code will be made publicly available.
-
deep neural nets with interpolating function as Output Activation
Neural Information Processing Systems, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and the code is available at https://github.com/ BaoWangMath/DNN-DataDependentActivation.
Xiyang Luo - One of the best experts on this subject based on the ideXlab platform.
-
Deep Neural Nets with Interpolating Function as Output Activation
arXiv: Learning, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and code will be made publicly available.
-
deep neural nets with interpolating function as Output Activation
Neural Information Processing Systems, 2018Co-Authors: Bao Wang, Xiyang Luo, Wei Zhu, Zuoqiang Shi, Stanley OsherAbstract:We replace the Output layer of deep neural nets, typically the softmax function, by a novel interpolating function. And we propose end-to-end training and testing algorithms for this new architecture. Compared to classical neural nets with softmax function as Output Activation, the surrogate with interpolating function as Output Activation combines advantages of both deep and manifold learning. The new framework demonstrates the following major advantages: First, it is better applicable to the case with insufficient training data. Second, it significantly improves the generalization accuracy on a wide variety of networks. The algorithm is implemented in PyTorch, and the code is available at https://github.com/ BaoWangMath/DNN-DataDependentActivation.