Explaining NonLinear Classification Decisions with Deep Taylor Decomposition
This addresses the interpretability problem for practitioners using DNNs in applications like image classification, though it is incremental as it builds on existing decomposition techniques.
The paper tackles the lack of transparency in deep neural networks by introducing a novel methodology for interpreting classification decisions through deep Taylor decomposition, which decomposes decisions into input contributions and is evaluated on MNIST and ILSVRC datasets.
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.