Regularizing neural network classification using random perturbations (A09)

通过训练的深度神经网络进行分类往往对输入上的对抗性噪声非常敏感。我们将研究几种提高深度学习模型稳健性的方法，包括随机平滑、网络参数的随机性和训练过程中对参数的约束。我们的目标是确保数学稳健性。此外，我们将把Pi Tempone最近引入的一种新的随机梯度下降的变种(多迭代随机估计)推广到训练中，并分析其收敛性质。

Classification via trained deep neural networks is often very sensitive to adversarial noise on the input. We will investigate several approaches for increasing the robustness of deep learning models including randomized smoothing, randomness on the network parameters and constraints on the parameters during training. We aim at mathematical robustness guarantees. Furthermore, we will extend a new variant of stochastic gradient descent (multiiteration stochastic estimator) recently introduced by PI Tempone for the training and will analyze its convergence properties.