Advances and Applications in Graph Neural Networks (GNNs)

Dear Colleagues,

Graph Neural Networks (GNNs) have garnered significant attention in machine learning research, enabling novel approaches to modeling and reasoning over complex relational data. This Special Issue seeks to explore both methodological advances in, and real‐world deployments of, GNNs, highlighting cutting‐edge innovations that are pushing the boundaries of what graph‐based AI can achieve.

We invite original contributions that address, but are not limited to, the following topics:

Core GNN architectures and optimization:

• Novel GNN architectures;
• Techniques to mitigate over-smoothing and graph bottlenecks;
• Scalable training strategies for billion-node graphs;

Transfer and few-shot learning on graphs:

• Domain adaptation across heterogeneous graph domains;
• Meta-learning frameworks for rapid adaptation to new topologies.

Dynamic, temporal, and spatio-temporal graphs:

• Modeling evolving network structures and streaming edge updates;
• Forecasting on spatio-temporal graphs (e.g., traffic, epidemic spread);
• Online learning and real-time inference in rapidly changing environments.

Graph explainability, interpretability, and trustworthiness:

• Counterfactual explanations and attribution methods for graph predictions;
• Uncertainty quantification, robustness to adversarial attacks, and certification.

Applications across domains:

• Bioinformatics and Cheminformatics: Protein–protein interactions, drug discovery, and genomics;
• Social Networks and Recommendation Systems: Influence maximization and user profiling;
• Computer Vision and Graphics: Scene graphs, point cloud analysis, and 3D shape generation;
• Natural Language Processing: Semantic dependency parsing and document‐level graphs;
• Finance and Economics: Fraud detection, knowledge graph risk analysis, and market modeling;
• Energy and Transportation: Power‐grid stability, traffic flow optimization, and route planning. 

Submissions that combine rigorous theoretical insights with compelling empirical validations—especially on large-scale real-world datasets—are highly encouraged. We also welcome comprehensive surveys that map the evolving landscape of graph neural methods.

Dr. Alessandro Rozza
Dr. Marco Leonardi
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. AI is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• graph neural networks
• transfer learning in GNNs
• GNN applications