The paper “EMMS: Multi-Label Multi-Dimensional Selection”, co-authored by Ph.D. student Li Yang, Prof. Yanyong Huang, and Prof. Jinyuan Chang of the team, CTO Ou Zheng of Zhiling Tech, Postdoctoral Fellow Minbo Ma of Tsinghua University, and Prof. Xiaoyi Jiang of the University of Münster, has been officially accepted by IJCAI 2026, one of the renowned international conferences in the field of Artificial Intelligence.
Abstract
Multi-label data are widely encountered in fields such as image recognition, text classification, and bioinformatics. Unlike traditional single-label data, each instance in a multi-label setting is typically associated with multiple labels. For example, a travel photo may be annotated with labels such as “beach”, “sunset” and “people”. However, real-world multi-label data often involve high dimensionality, outliers, and label noise. These issues can easily result in the curse of dimensionality and interfere with a model’s ability to learn from informative samples and reliable label information, thereby impairing the performance of downstream tasks such as classification and prediction.
To address the above problems, conventional methods usually select representative features, samples, or labels independently, while overlooking the interdependencies among them during the selection process. Moreover, these methods often assume that label annotations are noise-free, an assumption that is rarely valid in practical applications. To overcome these limitations, this paper proposes an evidence-theory-based multi-dimensional selection method for multi-label data, which jointly selects features, samples, and labels. Specifically, the proposed method employs a dual-projection framework with sparse constraints to map high-dimensional data first into a latent space and then into the label space. By explicitly modeling projection residuals, it identifies representative samples and thereby enables the joint selection of features, samples, and labels. Furthermore, evidence theory is introduced to integrate information from both the sample level and the label level, thereby improving the reliability of label learning and mitigating the adverse effects of noisy labels.
Extensive experimental results demonstrate that the proposed method outperforms existing state-of-the-art methods across multiple evaluation metrics. Case studies on multi-label image datasets further confirm its effectiveness.
Author Introduction
Jinyuan Chang is the Executive Director of the Joint Laboratory of Data Science and Business Intelligence at Southwestern University of Finance and Economics. He is a Guanghua Chair Professor and a recipient of the National Science Fund for Distinguished Young Scholars of China. He primarily engaged in research related to complex data analysis.
Li Yang is a Ph.D. student (Class of 2024) at Southwestern University of Finance and Economics, supervised by Prof. Yanyong Huang. She primarily engaged in research related to data mining and pattern recognition.
Minbo Ma is a Postdoctoral Fellow at Tsinghua University. He primarily engaged in interdisciplinary research related to the application of artificial intelligence and spatiotemporal data mining techniques in renewable energy and urban computing.
Xiaoyi Jiang is a Professor and Doctoral Supervisor at the Department of Mathematics and Computer Science, University of Münster. He is an IAPR Fellow, an AAIA Fellow, the Editor-in-Chief of the International Journal of Pattern Recognition and Artificial Intelligence, the Chairman of the German Chinese Professors Association, and the Chair of the International Workshop on Biomedical Imaging. He primarily engaged in research related to computer vision, pattern recognition, machine learning, and biomedical image analysis.
