Making CNNs Interpretable by Building Dynamic Sequential Decision Forests with Top-down Hierarchy Learning

TL;DR

This paper introduces a method to enhance CNN interpretability by integrating a differentiable decision forest that learns hierarchical decision structures guided by category semantics, enabling sequential decision-making for better explanations.

Contribution

The paper presents a novel approach to make CNNs interpretable by building a dynamic decision forest with top-down hierarchy learning, maintaining accuracy while improving interpretability.

Findings

dDSDF achieves higher classification accuracy than original CNNs.

The model produces more plausible hierarchies and saliency maps.

It enables sequential decision-making aligned with category semantics.

Abstract

In this paper, we propose a generic model transfer scheme to make Convlutional Neural Networks (CNNs) interpretable, while maintaining their high classification accuracy. We achieve this by building a differentiable decision forest on top of CNNs, which enjoys two characteristics: 1) During training, the tree hierarchies of the forest are learned in a top-down manner under the guidance from the category semantics embedded in the pre-trained CNN weights; 2) During inference, a single decision tree is dynamically selected from the forest for each input sample, enabling the transferred model to make sequential decisions corresponding to the attributes shared by semantically-similar categories, rather than directly performing flat classification. We name the transferred model deep Dynamic Sequential Decision Forest (dDSDF). Experimental results show that dDSDF not only achieves higher classification accuracy than its conuterpart, i.e., the original CNN, but has much better interpretability, as qualitatively it has plausible hierarchies and quantitatively it leads to more precise saliency maps.