arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2009.06571cs.LGstat.ML
收藏

神经网络的输入Hessian正则化

Input Hessian Regularization of Neural Networks

TU Kaiserslautern(凯泽斯劳滕工业大学);TU Berlin(柏林工业大学)

Waleed Mustafa, Robert A. Vandermeulen, Marius Kloft

摘要:

正则化的输入梯度已被证明是有效的,在提高神经网络的鲁棒性。因此,输入的Hessian的正则化是自然的下一步。这里的一个关键挑战是计算复杂性。计算输入的Hessian在计算上是不可行的。在本文中,我们提出了一个有效的算法训练深度神经网络与海森算子范数正则化。我们从理论上分析了这种方法,并证明了Hessian算子范数与神经网络抵御对抗攻击的能力有关。我们给出了初步的实验评估MNIST和FMNIST数据集,这表明,新的正则化可以,确实是可行的,而且,它增加了神经网络的鲁棒性输入梯度正则化。

英文摘要:

Regularizing the input gradient has shown to be effective in promoting the robustness of neural networks. The regularization of the input's Hessian is therefore a natural next step. A key challenge here is the computational complexity. Computing the Hessian of inputs is computationally infeasible. In this paper we propose an efficient algorithm to train deep neural networks with Hessian operator-norm regularization. We analyze the approach theoretically and prove that the Hessian operator norm relates to the ability of a neural network to withstand an adversarial attack. We give a preliminary experimental evaluation on the MNIST and FMNIST datasets, which demonstrates that the new regularizer can, indeed, be feasible and, furthermore, that it increases the robustness of neural networks over input gradient regularization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑