Explaining NonLinear Classification Decisions with Deep Taylor Decomposition
Grégoire Montavon∗, Sebastian Bach, Alexander Binder, Wojciech Samek∗, and Klaus-Robert Müller∗ This work was supported by the Brain Korea 21 Plus Program through the National Research Foundation of Korea funded by the Ministry of Education. This work was also supported by the grant DFG (MU 987/17-1) and by the German Ministry for Education and Research as Berlin Big Data Center BBDC (01IS14013A). This publication only reflects the authors views. Funding agencies are not liable for any use that may be made of the information contained herein. Asterisks indicate corresponding author.∗G. Montavon is with the Berlin Institute of Technology (TU Berlin), 10587 Berlin, Germany. (e-mail: Bach is with Fraunhofer Heinrich Hertz Institute, 10587 Berlin, Germany. (e-mail: Binder is with the Singapore University of Technology and Design, 487372, Singapore. (e-mail: Samek is with Fraunhofer Heinrich Hertz Institute, 10587 Berlin, Germany. (e-mail: Müller is with the Berlin Institute of Technology (TU Berlin), 10587 Berlin, Germany, and also with the Department of Brain and Cognitive Engineering, Korea University, Seoul 136-713, Korea (e-mail:
Abstract
Nonlinear methods such as Deep Neural Networks (DNNs) are the gold standard for various challenging machine learning problems, e.g., image classification, natural language processing or human action 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 is based on deep Taylor decomposition and 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.
中文速览
深度神经网络（Deep Neural Networks）在图像分类等任务上表现出色，却像一个"黑盒子"——人们只知道它给出了答案，却不知道它为什么这么判断。为了解决这一可解释性难题，作者提出了"深度泰勒分解"（Deep Taylor Decomposition）方法：将网络的最终分类得分，沿着网络结构从输出层一路反向拆解到输入层，把每个像素对最终决策的"贡献值"（相关性得分）都算出来，最终生成一张直观的热力图，高亮显示哪些像素真正驱动了模型的判断。该方法的关键创新在于把原本难以处理的整体泰勒展开，拆成每个神经元各自独立的局部展开，既有严谨的数学推导支撑，又能高效复用反向传播的网络结构，无需重新训练模型。在 MNIST 手写数字和 ILSVRC 大规模图像数据集上的实验表明，该方法生成的热力图能清晰、准确地定位模型决策所依赖的图像区域，为理解和验证深度神经网络的"推理过程"提供了坚实的理论基础和实用工具。
原文 arXiv:1512.02479;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1512.02479v1