Luca Virgili

Luca Virgili

Researcher Tenure-Track at Università Politecnica delle Marche

Home

I’m Luca Virgili, tenure-track researcher at the Department of Information Engineering, Università Politecnica delle Marche (Italy).

My research lies at the intersection of Artificial Intelligence and Complex Networks. I am particularly interested in understanding the behavior of interconnected systems and in developing machine learning models that are both interpretable and computationally efficient for real-world applications.

Research Interests

  • Complex and Social Networks analysis and modeling of online interactions
  • Graph Neural Networks (GNNs) and multilayer network models
  • Explainable AI (XAI) in computer vision and social contexts
  • Efficient deep learning models, including model pruning, quantization, and knowledge distillation
  • Large Language Models (LLMs) applied across domains such as education, and content generation

My work is often interdisciplinary, spanning social media, education, cybersecurity, and human-centered AI.

Projects

Below are some research highlights with potential real-world applications:

📊 Interpretable Neural Networks

We modeled convolutional neural networks (e.g., VGG16, ResNet50) as multilayer networks to understand and compress their structure.
Applications: Model compression, visual explainability.

🔍 Social Influence and Behavior Prediction on Online Platforms

We used GNNs to detect key influencers and community evolution patterns in social media networks.
Applications: Marketing strategy, content moderation, misinformation detection.

🧠 Efficient Vision Transformers via Pruning and Knowledge Distillation

Our approach improves the performance and interpretability of ViTs while reducing complexity.
Applications: Embedded AI, model compression, visual explainability.

Want to explore collaboration opportunities? Get in touch!

Teaching

I teach undergraduate and graduate courses on:

  • Fundamentals of Computer Science
  • Object-Oriented Programming
  • Big Data Labs (Qlik, Tableau, PowerBI)
  • Data Science Labs (Python, Pandas, ML)
  • AI-Driven Cybersecurity

See the Teaching section for materials and syllabi.

Thesis Opportunities

If you’re interested in a thesis on:

  • Explainable AI
  • Complex Network Analysis
  • Vision Transformers
  • Applications of GNNs

Check the Thesis Proposal page.

Contact

I’m available for academic collaborations, industry projects, and supervision.