Graph Filters for Signal Processing and Machine Learning on Graphs

Elvin Isufi,Fernando Gama,David I. Shuman,Santiago Segarra

2024-02-20

Abstract:Filters are fundamental in extracting information from data. For time series and image data that reside on Euclidean domains, filters are the crux of many signal processing and machine learning techniques, including convolutional neural networks. Increasingly, modern data also reside on networks and other irregular domains whose structure is better captured by a graph. To process and learn from such data, graph filters account for the structure of the underlying data domain. In this article, we provide a comprehensive overview of graph filters, including the different filtering categories, design strategies for each type, and trade-offs between different types of graph filters. We discuss how to extend graph filters into filter banks and graph neural networks to enhance the representational power; that is, to model a broader variety of signal classes, data patterns, and relationships. We also showcase the fundamental role of graph filters in signal processing and machine learning applications. Our aim is that this article provides a unifying framework for both beginner and experienced researchers, as well as a common understanding that promotes collaborations at the intersections of signal processing, machine learning, and application domains.

Signal Processing,Machine Learning

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to use graph filters to extract information when processing and learning graph - structured data. Traditional signal processing methods are suitable for data in Euclidean domains such as time series and images, but these methods are not applicable to modern data on networks and other irregular domains. Graph filters can take into account the structural characteristics of the data domain, thus effectively processing and learning such data. The paper provides a comprehensive overview of graph filters, including different types of filters, design strategies, and their applications in signal processing and machine learning.

Specifically, the paper focuses on the following aspects:
1. **Basic concepts of graph filters**: Introduce the definition, properties of graph filters, and their differences and connections with traditional filters.
2. **Design and implementation of graph filters**: Discuss how to design different types of graph filters and the implementation methods of these filters in practical applications.
3. **Extensions of graph filters**: Explore how to extend graph filters to filter banks and graph neural networks to enhance their representational capabilities.
4. **Applications of graph filters**: Demonstrate the specific applications of graph filters in signal processing and machine learning, including signal reconstruction, anomaly detection, image processing, distributed signal processing, semi - supervised and unsupervised learning, matrix completion, Gaussian process regression, point cloud processing, and computer vision.

Through these contents, the paper aims to provide a unified framework for beginners and experienced researchers, promoting cooperation in the fields of signal processing, machine learning, and applications.