Valerio Di Pasquale

Valerio Di Pasquale

PhD Student in Computer Science and Information Technology at Università degli Studi di Salerno, focusing on artificial intelligence for complex networks, graphs, hypergraphs, and structured-data reasoning.

About Me

I am a PhD Student in Computer Science and Information Technology at the Università degli Studi di Salerno. My work focuses on artificial intelligence for complex networks, with particular attention to graph and hypergraph generative models, link and hyperlink prediction, motif analysis, and graph-based representation learning.

I am especially interested in combining graph models with large language models to support graph-based knowledge extraction, reasoning over structured data, and more interpretable analysis of complex systems. Alongside research, I have professional experience in web design and development and a long-standing freelance background building applications for individuals and small businesses.

I am based in Salerno, Italy, and I currently work across research, software development, and academic projects at the intersection of machine learning and network science.

Research

My primary research interests revolve around the study and application of artificial intelligence algorithms for complex networks. I work on graphs and hypergraphs, studying how generative and predictive models can capture the structure and dynamics of interconnected systems in domains such as social networks, biology, and communication systems.

Current Directions

  • Generative models for graphs and hypergraphs
  • Link prediction and hyperlink prediction
  • Motif analysis in real-world networks
  • Graph representation learning
  • Geometric deep learning and graph neural networks

Broader Interests

  • Large language models for structured data reasoning
  • Graph-to-text generation and question answering over graphs
  • Diffusion models, GANs, and variational autoencoders
  • Applications to biology, social systems, and computational design
  • Interpretable AI methods for network analysis

Selected Publication

  • Alessia Antelmi, Gennaro Cordasco, Daniele De Vinco, Valerio Di Pasquale, Mirko Polato, and Carmine Spagnuolo. 2025. Hypergraph Motif Representation Learning. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '25), 13-24.

Education

  • PhD Student, Computer Science and Information Technology @ Università degli Studi di Salerno
    November 2025 - Present
  • Master's Degree in Computer Science @ Università degli Studi di Salerno
    October 2023 - July 2025
    GPA: 4.0/4.0 (29.75/30). Main coursework included Cloud Computing, Social Networks, Distributed Architectures, Blockchain, Data Science, and Parallel and Distributed Programming.
  • Excellence Program (Percorso di Eccellenza) @ Università degli Studi di Salerno
    2024 - 2025
    Selected among 18 students of the Computer Science Department for outstanding academic achievements and for a research-oriented advanced pathway.
  • 8th Advanced Course on Data Science & Machine Learning @ Castiglione della Pescaia, Tuscany
    9 June 2025 - 12 June 2025
    Presented research work as both an oral presentation and a poster. Attendance above 85% was recognized with 8 ECTS credits.
  • Bachelor's Degree in Computer Science @ Università degli Studi di Salerno
    September 2020 - October 2023
    GPA: 3.925/4.0 (28.9/30). Graduation mark: 110/110 cum laude, with a special mention from the commission.
    Dissertation: Generative Models for Hypergraphs using Deep Learning.

Experience

Research Activities

Professional Experience

  • Web Design and Development @ Centro Regionale Information Communication Technology
    February 2024 - May 2024
    Contributed to the 3Dify project, focusing on frontend design and on best practices for building a web application for automatically turning 2D human images into animated customizable avatars with emotions.
  • Freelance Developer
    Since 2017
    Worked with private individuals and small-to-medium businesses across Italy, mainly using Angular, Django, and WordPress for web development projects.

Projects

  • Higher-Order Networks
    A curated collection of hypergraph datasets stored in HIF (Hypergraph Interchange Format), including precomputed node and hyperedge features. Published on Hugging Face Hub.
  • Health Tracking System
    A cloud-based system with a web frontend and a serverless backend for tracking health data from IoT devices, designed as a scalable and low-cost architecture.
  • Parallel Deep Generative Model for Hypergraph
    Bachelor's thesis project focused on a deep variational autoencoder for generating synthetic hypergraphs in parallel, implemented in Python with PyTorch.
  • Dubito
    A Python package for monitoring listings on subito.it through queries and filters, developed as a practical public tool from a personal need.

Skills

My technical work combines machine learning research, graph analysis, high-performance computing, and web development. Python is my primary language for research, while C/C++, Cython, and modern web technologies support production, optimization, and applied development work.

AI & Research

  • PyTorch
  • PyTorch Geometric
  • PyTorch Lightning
  • Scikit-Learn
  • Graph and hypergraph analysis

Programming

  • Python
  • C
  • C++
  • Cython
  • NumPy

Web Development

  • Angular
  • Django
  • WordPress
  • Frontend architecture
  • SEO-oriented content management

Contact

You can reach me at valeriodipasq@gmail.com or v.dipasquale3@studenti.unisa.it. I am always happy to discuss research collaborations, academic opportunities, and software projects.

You can also connect with me on LinkedIn and explore my code and public repositories on GitHub.