Multi-View Non-negative Matrix Factorization Discriminant Learning via Cross Entropy Loss

TL;DR

This paper introduces a novel multi-view learning method that enhances discriminative feature extraction using joint non-negative matrix factorization combined with cross entropy loss, leading to improved classification performance.

Contribution

It proposes an improved algorithm that incorporates cross entropy loss into joint non-negative matrix factorization for better discriminative feature learning in multi-view data.

Findings

Achieves better classification accuracy than the original algorithm.

Demonstrates superiority over several state-of-the-art methods.

Validates effectiveness on multiple datasets.

Abstract

Multi-view learning accomplishes the task objectives of classification by leverag-ing the relationships between different views of the same object. Most existing methods usually focus on consistency and complementarity between multiple views. But not all of this information is useful for classification tasks. Instead, it is the specific discriminating information that plays an important role. Zhong Zhang et al. explore the discriminative and non-discriminative information exist-ing in common and view-specific parts among different views via joint non-negative matrix factorization. In this paper, we improve this algorithm on this ba-sis by using the cross entropy loss function to constrain the objective function better. At last, we implement better classification effect than original on the same data sets and show its superiority over many state-of-the-art algorithms.