Embedding Semantic Hierarchy in Discrete Optimal Transport for Risk Minimization

TL;DR

This paper introduces a risk-aware discrete optimal transport framework that embeds semantic hierarchical information into training, improving classification by considering semantic distances rather than just label accuracy.

Contribution

It proposes a novel method to incorporate semantic hierarchy into optimal transport for risk minimization in classification tasks.

Findings

Improved classification accuracy on large-scale image datasets.

Effective integration of semantic hierarchy into the training process.

Plug-and-play applicability to existing models.

Abstract

The widely-used cross-entropy (CE) loss-based deep networks achieved significant progress w.r.t. the classification accuracy. However, the CE loss can essentially ignore the risk of misclassification which is usually measured by the distance between the prediction and label in a semantic hierarchical tree. In this paper, we propose to incorporate the risk-aware inter-class correlation in a discrete optimal transport (DOT) training framework by configuring its ground distance matrix. The ground distance matrix can be pre-defined following a priori of hierarchical semantic risk. Specifically, we define the tree induced error (TIE) on a hierarchical semantic tree and extend it to its increasing function from the optimization perspective. The semantic similarity in each level of a tree is integrated with the information gain. We achieve promising results on several large scale image classification tasks with a semantic tree structure in a plug and play manner.