Explaining the Unexplained: A CLass-Enhanced Attentive Response (CLEAR) Approach to Understanding Deep Neural Networks

2017/04/13 by Devinder Kumar, Kumar, Devinder, Alexander Wong +3
Computer Science · #Artificial Intelligence (cs.AI) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multimedia (cs.MM) #cs.AI #cs.CV #cs.LG #cs.MM

paper · pdf · doi:10.48550/arxiv.1704.04133

Accepted at Computer Vision and Patter Recognition Workshop (CVPR-W) on Explainable Computer Vision, 2017

arxiv created 2017/05/18 · arxiv updated 2017/05/22

Abstract

In this work, we propose CLass-Enhanced Attentive Response (CLEAR): an approach to visualize and understand the decisions made by deep neural networks (DNNs) given a specific input. CLEAR facilitates the visualization of attentive regions and levels of interest of DNNs during the decision-making process. It also enables the visualization of the most dominant classes associated with these attentive regions of interest. As such, CLEAR can mitigate some of the shortcomings of heatmap-based methods associated with decision ambiguity, and allows for better insights into the decision-making process of DNNs. Quantitative and qualitative experiments across three different datasets demonstrate the efficacy of CLEAR for gaining a better understanding of the inner workings of DNNs during the decision-making process.

Related

• A New Method to Visualize Deep Neural Networks
• A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks
• Neural Networks Decoded: Targeted and Robust Analysis of Neural Network Decisions via Causal Explanations and Reasoning
• NeuroView: Explainable Deep Network Decision Making
• Explaining Deep Neural Networks by Leveraging Intrinsic Methods