Explaining nonlinear classification decisions with deep Taylor decomposition
Explaining nonlinear classification decisions with deep Taylor decomposition
复制标题
DOI：
10.1016/j.patcog.2016.11.008
复制
发表时间：
2017-05-01
影响因子：
8
通讯作者：
Mueller, Klaus-Robert
中科院分区：
文献类型：
--
作者：
Montavon, Gregoire;Lapuschkin, Sebastian;Mueller, Klaus-Robert
深度神经网络（DNN）等非线性方法是图像识别等各种具有挑战性的机器学习问题的黄金标准。虽然这些方法表现得令人印象深刻，但它们有一个显著的缺点，即缺乏透明度，限制了解决方案的可解释性，从而限制了实际应用的范围。特别是DNN由于其多层非线性结构而充当黑匣子。在本文中，我们介绍了一种新的方法来解释通用多层神经网络的网络分类决策分解成其输入元素的贡献。虽然我们的重点是图像分类，但该方法适用于广泛的输入数据，学习任务和网络架构。我们的方法称为深度泰勒分解，通过将解释从输出层反向传播到输入层，有效地利用了网络的结构。我们评估所提出的方法经验上的MNIST和ILSVRC数据集。
Nonlinear methods such as Deep Neural Networks (DNNs) are the gold standard for various challenging machine learning problems such as image recognition. Although these methods perform impressively well, they have a significant disadvantage, the lack of transparency, limiting the interpretability of the solution and thus the scope of application in practice. Especially DNNs act as black boxes due to their multilayer nonlinear structure. In this paper we introduce a novel methodology for interpreting generic multilayer neural networks by decomposing the network classification decision into contributions of its input elements. Although our focus is on image classification, the method is applicable to a broad set of input data, learning tasks and network architectures. Our method called deep Taylor decomposition efficiently utilizes the structure of the network by backpropagating the explanations from the output to the input layer. We evaluate the proposed method empirically on the MNIST and ILSVRC data sets.