Scholar
Fabrizio Frasca
Google Scholar ID: PT2CDA4AAAAJ
Posdoctoral Fellow, Technion – Israel Institute of Technology
Citations & Impact
Citations
4,598
H-index
13
i10-index
15
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Multiple papers accepted at international conferences such as ICML and ECML-PKDD; involved in several research projects, including Subgraph GNNs and Laplacian Positional Encodings for Temporal GNNs.
Research Experience
- Machine Learning Researcher at Twitter Cortex from 2019 – acquisition of Fabula AI – to early 2023. Currently, a Postdoctoral Fellow at Technion, working with Prof. Haggai Maron on Geometric Deep Learning.
Education
- PhD in Computing from Imperial College London, supervised by Prof. Michael Bronstein, with research focusing on overcoming the intrinsic representational limits of GNNs.
Background
- Research interests revolve around principled and effective learning over structured data, with a particular emphasis on equivariance and expressiveness. Recently, he has been working on architectures to learn from (structured) computational traces of Large Language Models (LLMs) for the automated detection of problematic behavioral patterns such as hallucinations. He also extensively works on methods for learning on graphs, focusing on designing efficient and expressive Graph Neural Networks (GNNs).
Miscellany
- One of the co-founders and co-organisers of the GLOW (Graph Learning on Wednesdays) reading group.
Co-authors