Visualizing the Decision-making Process in Deep Neural Decision Forest

Shichao Li, Kwang-Ting Cheng; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2019, pp. 114-117

Abstract


Deep neural decision forest (NDF) achieved remarkable performance on various vision tasks via combining decision tree and deep representation learning. In this work, we first trace the decision-making process of this model and visualize saliency maps to understand which portion of the input influence it more for both classification and regression problems. We then apply NDF on a multi-task coordinate regression problem and demonstrate the distribution of routing probabilities, which is vital for interpreting NDF yet not shown for regression problems. The pre-trained model and code for visualization will be available at https://github.com/Nicholasli1995/ VisualizingNDF

Related Material


[pdf] [dataset]
[bibtex]
@InProceedings{Li_2019_CVPR_Workshops,
author = {Li, Shichao and Cheng, Kwang-Ting},
title = {Visualizing the Decision-making Process in Deep Neural Decision Forest},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2019}
}