Stabilizing Adversarial Invariance Induction from Divergence Minimization Perspective-猫眼课题宝

DOI：

10.24963/ijcai.2020/271

发表时间：

2020-07

期刊：

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Yusuke Iwasawa;K. Akuzawa;Y. Matsuo

对抗不变性归纳（AII）是一个通用而强大的框架，用于将滋扰属性的不变性强制执行到神经网络表示中。然而，它的优化往往是不稳定的，很少有人知道它的实际行为。本文提出了一个优化困难的原因进行了分析，并提供了一个更好的优化过程，从发散最小化的角度重新思考AII。有趣的是，这种观点指出了优化困难的原因：它不能确保适当的发散最小化，这是不变表示的要求。然后，我们提出了一个简单的变体AII，称为不变性感应匹配，它考虑到了发散最小化的不变表示的解释。我们的方法在具有各种配置的玩具数据集中始终实现了近最优的不变性，其中原始AII是灾难性的不稳定。四个真实世界的数据集上的扩展实验也支持所提出的方法的上级性能，从而提高用户匿名和域泛化。

Adversarial invariance induction (AII) is a generic and powerful framework for enforcing an invariance to nuisance attributes into neural network representations. However, its optimization is often unstable and little is known about its practical behavior. This paper presents an analysis of the reasons for the optimization difficulties and provides a better optimization procedure by rethinking AII from a divergence minimization perspective. Interestingly, this perspective indicates a cause of the optimization difficulties: it does not ensure proper divergence minimization, which is a requirement of the invariant representations. We then propose a simple variant of AII, called invariance induction by discriminator matching, which takes into account the divergence minimization interpretation of the invariant representations. Our method consistently achieves near-optimal invariance in toy datasets with various configurations in which the original AII is catastrophically unstable. Extentive experiments on four real-world datasets also support the superior performance of the proposed method, leading to improved user anonymization and domain generalization.

Stabilizing Adversarial Invariance Induction from Divergence Minimization Perspective

文献摘要

相似文献