Andrea Pollastro

Hey there! I am a Postdoctoral Researcher at University of Naples “Federico II” (Naples, IT). My research focuses on developing, applying and improving Machine Learning methodologies, with a particular focus on Artificial Neural Networks.

My academic background is in Computer Science: I graduated in 2019 with a MS in Computer Science and in 2016 with a BS in Computer Science, both from the University of Naples “Federico II”. In 2024, I completed my PhD in Information and Communication Technology for Health at the same institution focusing on Deep Learning methods.

In 2023, I was awarded a scholarship as a visiting PhD student at the Lawrence Berkeley National Laboratory (Berkeley, CA, USA) to work on applications of Machine Learning techniques on data gathered during the operation of the third-generation synchrotron light source Advanced Light Source (ALS) in collaboration with the ALS Accelerator Physics Group. I am currently a research affiliate at the same laboratory.

In 2024, I joined CERN (Geneva, CH) as a research affiliate at the CLEAR facility, focusing on exploring the application of Machine Learning techniques to forecasting problems for particle accelerator research. In the same year, as a visiting postdoctoral researcher, I was the Principal Investigator during an experimental week (10–16 November), funded by the European Union’s “Horizon Europe - Research and Innovation” programme. The focus of this week was to evaluate the CLEAR accelerator as an AI-driven testbed for CERN’s Efficient Particle Accelerator (EPA) project.

In 2024, I co-authored the paper “PhysBERT: A Text Embedding Model for Physics Scientific Literature”, which was accepted at the Machine Learning and the Physical Sciences workshop at the NeurIPS 2024 conference. This work introduced PhysBERT, the first sentence embedding model specifically designed for the physics domain. Both the model and the dataset are available on HuggingFace.