Matteo Biagetti

Matteo Biagetti

Researcher

AREA Science Park

About me

I am a research scientist. After working for several years in theoretical cosmology, I’ve recently moved towards data-oriented research. I’m using Topological Data Analysis techniques to study the geometry of representations of convolutional neural networks and large transformer models applied to a diverse range of datasets, including protein sequences, images, and text. I am also the scientific manager of a project funded by the Italian government called FAIR-by-design, to develop FAIR workflows of data and metadata for laboratories of material and life sciences.

I am currently employed as a Data Scientist at the Laboratory of Data Engineering (LADE) at the Institute of Research and Technological Innovation, Area Science Park in Trieste, Italy.

Interests
  • Topological Data Analysis
  • Machine Learning
  • FAIR-by-design
  • Theoretical Physics
Education
  • PhD in Theoretical Physics, 2016

    University of Geneva

  • MSc in Theoretical Physics, 2011

    University of Padova

  • BSc in Physics, 2009

    University of Padova

Experience

 
 
 
 
 
Researcher
AREA Science Park
June 2023 – Present Trieste - Italy
 
 
 
 
 
Postdoctoral fellow
International School for Advanced Studies - SISSA
September 2020 – June 2023 Trieste - Italy
 
 
 
 
 
Postdoctoral fellow
University of Amsterdam
October 2016 – August 2020 Amsterdam - The Netherlands

Projects

*
EOS
Eos is known as the goddess of dawn in Greek mythology. The Eos Dataset is a set of numerical Nbody simulations of the Universe, investigating the cosmological evolution of primordial interactions taking place at the origin of the Universe.
EOS
FAIR-by-design
A project for the FAIRification and the implementation of FAIR-by-design workflows in all experiments lead and co-lead by the Institute of Research and Innovation Technology at AREA Science Park.
FAIR-by-design
PHOCUS
PHOCUS: Persistent Homology Of Cosmic strUctreS is an ambitious project comprising of senior and junior researchers as well as PhD candidates and master students to apply techniques drawn from topological data analysis, specifically persistent homology, to cosmological datasets.
PHOCUS

Recent & Upcoming Talks

Contact