Senior Researcher
phone: +39 091 6809 464 (direct)
email: antonio.pagliaro@inaf.it
Main scientific interests: Artificial Intelligence, Machine and Deep Learning, Explainable AI, Computer Vision, High Energy Astrophysics, Data Driven Finance.
Principal Investigator of the MADELEinE research program. Local Manager for the ICSC National Centre for HPC, Big Data and Quantum Computing. Co-PI of CRUNCH.
Academic Editor and Section Board Member at Applied Sciences.
Projects: HPC, Big Data and Quantum Computing, MADELEinE, ASTRI, CTA, JEM-EUSO, CRUNCH, HERMES, Muography
Selected Recent Publications
Book Chapters
- Time Series Analysis in Machine Learning — in Machine Learning Techniques for Astrophysics and Cosmology (eds. Bambi, Kashyap, Shashank, Yoshida), Springer Singapore, in press (2027)
- Machine Learning for Event Reconstruction in Imaging Atmospheric Cherenkov Telescopes — in Machine Learning Techniques for Astrophysics and Cosmology (eds. Bambi, Kashyap, Shashank, Yoshida), Springer Singapore, in press (2027)
- Application of Machine and Deep Learning Methods to the Analysis of IACTs Data — Springer (ISBN 978-3-030-65867-0), 2021
High Energy Astrophysics
Data Driven Finance
- Regime-Aware LightGBM for Stock Market Forecasting: A Validated Walk-Forward Framework with Statistical Rigor and Explainable AI Analysis — Electronics 15, 2026
- Cognitive Biases in Asset Pricing: An Empirical Analysis of the Alphabet Effect and Ticker Fluency in the US Market — Symmetry 18, 2026
- Artificial Intelligence vs. Efficient Markets: A Critical Reassessment of Predictive Models in the Big Data Era — Electronics 14, 2025
- An Introduction to Machine Learning Methods for Fraud Detection — Appl. Sci. 15, 2025
- Forecasting Significant Stock Market Price Changes Using Machine Learning: Extra Trees Classifier Leads — Electronics 12, 2023
Computer Vision, LLMs & More
- Cognitive Biases in Large Language Models: A Systematic Quantitative Assessment and Debiasing Analysis — Electronics 15, 2026
- The Specialization of Intelligence in AI Horizons: Present Status and Visions for the Next Era — Appl. Sci. 15, 2025
- Advanced AI and Machine Learning Techniques for Time Series Analysis and Pattern Recognition — Appl. Sci. 15, 2025 (editorial)
- An Introduction to Machine and Deep Learning Methods for Cloud Masking Applications — Appl. Sci. 14, 2024
- AI in Experiments: Present Status and Future Prospects — Appl. Sci. 13, 2023
Within MADELEinE we apply machine and deep learning to the analysis of large scientific datasets, mainly in high energy astrophysics and astroparticle physics, and in parallel to data-driven approaches to financial markets. A central thread of our work is explainable AI: a model is useful to science only insofar as its decisions can be understood, validated and trusted. We are interested in a few recurring questions — how a model should represent its input, how to tell a genuine pattern from a statistical fluctuation, how to make use of prior physical knowledge, and how to present results in a form that a domain expert can interpret and check.
Data Science and Artificial Intelligence
Our work in data science and artificial intelligence spans several connected areas:
- Machine and deep learning — supervised, unsupervised and semi-supervised methods for large-scale, high-dimensional scientific data.
- Explainable AI — quantitative assessment of model interpretability (faithfulness, complexity, plausibility), so that predictions can be examined rather than taken on trust.
- Large language models — systematic, quantitative evaluation of their behaviour and biases, with debiasing strategies applied at inference time.
- Computer vision — image and signal analysis, including segmentation and detection, applied to instrument and survey data.
- Predictive modeling and data mining — pattern extraction, classification and forecasting on complex datasets, including financial time series.
More: Data Science
MADELEinE
MADELEinE (MAchine and DEep LEarning in Experiments) is the research program I coordinate at INAF–IASF Palermo. We bring machine and deep learning methods to the large datasets produced by telescopes and particle detectors, across high energy astrophysics, astroparticle physics and quantitative finance, with attention to making the resulting models interpretable and reproducible.
More: MADELEinE
ASTRI
ASTRI (Astrofisica con Specchi a Tecnologia Replicante Italiana), led by INAF, develops small Imaging Atmospheric Cherenkov Telescopes with a wide field of view. Our contribution sits in data processing and software reconstruction: machine-learning methods for gamma/hadron separation and energy reconstruction, including the use of temporal pixel information alongside the classical morphological parameters, and ensemble approaches for event reconstruction.
More: ASTRI
JEM-EUSO
The JEM-EUSO program (Joint Experiment Missions for Extreme Universe Space Observatory) studies the nature and origin of ultra-high energy cosmic rays and neutrinos, using a super-wide-field telescope to detect the UV light from the air showers they produce in the atmosphere. The collaboration involves about 300 researchers from 16 countries; the IASF Palermo group contributes to the data analysis side.
More: JEM-EUSO