The New Special Issue "Statistical Machine Learning with High-Dimensional Data and Image Analysis: Second Edition" is Open for Submission!

Guest Editor: Dr. Lei Wang

Submit to the Special Issue: https://www.mdpi.com/journal/entropy/special_issues/YK9QV59905

Submission deadline: 30 November 2025

Special Issue Information:

Statistical machine learning methods have been widely used for the analysis of high-dimensional structured data and images. Many algorithms have been developed aiming to build models based on sample data from various areas, such as medicine, web documents, remote sensing, and multimedia data. Traditional techniques like sparse signal reconstruction, low-rank matrix recovery, and principal component analysis have been extensively studied to handle high-dimensional data. However, solving convex or non-convex optimization problems and developing theories for new kinds of structured data and images remain active research areas.

The applications of statistical machine learning to high-dimensional data often face difficulties such as non-modularity and instability, which limit their effectiveness in real-world scenarios. Emerging technologies, particularly deep neural networks, have provided new solutions for large-scale datasets. However, the interpretability of these networks is still not as good as that of traditional statistical machine learning algorithms. Moreover, the exploration of information entropy in deep learning is still in its early stages. For example, recent research has proposed new methods to analyze deep neural networks via information plane theory, but estimating mutual information in high-dimensional hidden layers remains challenging.

In addition, the rapid development of large vision–language models has introduced new opportunities and challenges. These models have shown great potential in various applications, but their integration with traditional machine learning methods and the improvement in their interpretability are still open questions. Furthermore, the latest advancements in statistical machine learning, such as model compression, federated learning, and data privacy, have also brought new perspectives to the field.

This Special Issue aims to be a forum for presenting new techniques of statistical machine learning for high-dimensional data. We particularly welcome contributions on the analysis and interpretation of real-world data or images based on machine learning, deep learning, or large vision–language models. Topics of interest include, but are not limited to the following:

• Advanced sparse signal reconstruction and low-rank matrix recovery techniques;
• New methods for solving convex and non-convex optimization problems in high-dimensional settings;
• Enhancing the interpretability of deep neural networks and large vision–language models;
• Exploring the application of information entropy in deep learning;
• Addressing non-modularity and instability issues in statistical machine learning;
• Innovative applications of statistical machine learning in real-world scenarios, such as healthcare, environmental monitoring, and smart manufacturing;
• Integration of traditional machine learning methods with large vision–language models;
• Contributions related to model compression, federated learning, and data privacy in the context of high-dimensional data.