DeepAI
Log In Sign Up

Understanding Individual Decisions of CNNs via Contrastive Backpropagation

12/05/2018
by   Jindong Gu, et al.
20

A number of backpropagation-based approaches such as DeConvNets, vanilla Gradient Visualization and Guided Backpropagation have been proposed to better understand individual decisions of deep convolutional neural networks. The saliency maps produced by them are proven to be non-discriminative. Recently, the Layer-wise Relevance Propagation (LRP) approach was proposed to explain the classification decisions of rectifier neural networks. In this work, we evaluate the discriminativeness of the generated explanations and analyze the theoretical foundation of LRP, i.e. Deep Taylor Decomposition. The experiments and analysis conclude that the explanations generated by LRP are not class-discriminative. Based on LRP, we propose Contrastive Layer-wise Relevance Propagation (CLRP), which is capable of producing instance-specific, class-discriminative, pixel-wise explanations. In the experiments, we use the CLRP to explain the decisions and understand the difference between neurons in individual classification decisions. We also evaluate the explanations quantitatively with a Pointing Game and an ablation study. Both qualitative and quantitative evaluations show that the CLRP generates better explanations than the LRP.

READ FULL TEXT

page 5

page 10

page 14

page 20

page 21

08/06/2019

Explaining Convolutional Neural Networks using Softmax Gradient Layer-wise Relevance Propagation

Convolutional Neural Networks (CNN) have become state-of-the-art in the ...
02/11/2018

Influence-Directed Explanations for Deep Convolutional Networks

We study the problem of explaining a rich class of behavioral properties...
03/16/2020

Towards Ground Truth Evaluation of Visual Explanations

Several methods have been proposed to explain the decisions of neural ne...
05/18/2018

A Theoretical Explanation for Perplexing Behaviors of Backpropagation-based Visualizations

Backpropagation-based visualizations have been proposed to interpret con...
12/20/2019

When Explanations Lie: Why Modified BP Attribution Fails

Modified backpropagation methods are a popular group of attribution meth...
05/29/2019

Generating Contrastive Explanations with Monotonic Attribute Functions

Explaining decisions of deep neural networks is a hot research topic wit...