An evidential classifier based on Dempster-Shafer theory and deep learning

TL;DR

This paper introduces an innovative classifier combining Dempster-Shafer theory with deep learning to enhance set-valued classification, improving accuracy and decision caution across various data types.

Contribution

It presents an end-to-end trainable deep-learning classifier that integrates Dempster-Shafer theory for set-valued classification, including a novel partial multi-class act selection method.

Findings

Improved classification accuracy on image and signal tasks.

Enhanced decision-making with cautious set assignments.

Effective integration of DS theory with CNN architecture.

Abstract

We propose a new classifier based on Dempster-Shafer (DS) theory and a convolutional neural network (CNN) architecture for set-valued classification. In this classifier, called the evidential deep-learning classifier, convolutional and pooling layers first extract high-dimensional features from input data. The features are then converted into mass functions and aggregated by Dempster's rule in a DS layer. Finally, an expected utility layer performs set-valued classification based on mass functions. We propose an end-to-end learning strategy for jointly updating the network parameters. Additionally, an approach for selecting partial multi-class acts is proposed. Experiments on image recognition, signal processing, and semantic-relationship classification tasks demonstrate that the proposed combination of deep CNN, DS layer, and expected utility layer makes it possible to improve classification accuracy and to make cautious decisions by assigning confusing patterns to multi-class sets.