Francesco Ferrini

prof_pic.png

Located in Trento, Italy

Hi there 🌱

I am Francesco Ferrini, a researcher in Artificial Intelligence and Machine Learning. I am currently pursuing a Ph.D. in Artificial Intelligence and Machine Learning at the Structured Machine Learning Group, University of Trento, Italy, under the supervision of Andrea Passerini (University of Trento) and Manfred Jaeger (University of Aalborg, Denmark).

My research focuses on Graph Neural Networks and relational learning, with particular interest in multi-relational and heterogeneous graph models, as well as learning paradigms that explicitly account for incomplete, uncertain, or imperfect information in real-world data.

During my Ph.D., I have also carried out a research period in Japan, where I worked at the National Institute of Advanced Industrial Science and Technology (AIST) in collaboration with academic and industrial researchers. This experience allowed me to further explore graph-based learning in applied and interdisciplinary settings, and to strengthen my international research profile.

Prior to my doctoral studies, I obtained a Master's degree in Artificial Intelligence Systems. During this period, I worked on several projects spanning Machine Learning, Deep Learning, Computer Vision, and Natural Language Processing.

I also interned at Aalborg University in Denmark, where I began my Master's thesis research on meta-path learning for Graph Neural Networks in heterogeneous graphs.

Outside of research, I enjoy practicing sports—especially calisthenics and tennis—and playing chess.

Feel free to explore my website to learn more about my research, projects, and publications. If you are interested in collaboration, have a research idea, or would simply like to discuss science and technology, do not hesitate to get in touch.

  • Contact me at francescoferrini1997@gmail.com

Publications

Rethinking GNNs and Missing Features: Challenges, Evaluation and a Robust Solution

Authors: Francesco Ferrini, Veronica Lachi, Antonio Longa, Bruno Lepri, Matono Akiyoshi, Andrea Passerini, Xin Liu, Manfred Jaeger

Type: arXiv Preprint

Year: 2026

Bridging Theory and Practice in Link Representation with Graph Neural Networks

Authors: Veronica Lachi*, Francesco Ferrini*, Antonio Longa, Bruno Lepri, Andrea Passerini, Manfred Jaeger

Conference: NeurIPS 2025

Accepted as: Spotlight

Year: 2025

GNNs Meet Sequence Models Along the Shortest Path: An Expressive Method for Link Prediction

Authors: Francesco Ferrini*, Veronica Lachi*, Antonio Longa, Bruno Lepri, Andrea Passerini

Workshop: NeurIPS 2025 – New Perspectives in Graph Machine Learning

Year: 2025

Beyond Sparse Benchmarks: Evaluating GNNs with Realistic Missing Features

Authors: Francesco Ferrini, Veronica Lachi, Antonio Longa, Bruno Lepri, Andrea Passerini, Xin Liu, Manfred Jaeger

Workshop: NeurIPS 2025 – New Perspectives in Graph Machine Learning

Accepted as: Oral

Year: 2025

A Self-Explainable Heterogeneous GNN for Relational Deep Learning

Authors: Francesco Ferrini, Antonio Longa, Andrea Passerini, Manfred Jaeger

Journal: Transactions on Machine Learning Research (TMLR)

Status: Accepted

Year: 2025

A Simple and Expressive Graph Neural Network-Based Method for Structural Link Representation

Authors: Veronica Lachi, Francesco Ferrini, Antonio Longa, Bruno Lepri, Andrea Passerini

Conference: Proceedings of Machine Learning Research (PMLR)

Year: 2024

Meta-Path Learning for Multi-Relational Graph Neural Networks

Authors: Francesco Ferrini, Antonio Longa, Andrea Passerini, Manfred Jaeger

Conference: Learning on Graphs Conference (LoG)

Accepted as: Oral

Year: 2023

Energy-Efficient Inference on the Edge Exploiting TinyML Capabilities for UAVs

Authors: Wamiq Raza, Anas Osman, Francesco Ferrini, Francesco De Natale

Journal: Drones (MDPI)

Year: 2021