Explainable Scientific Machine Learning: Theoretical and Practical Perspectives
SPEAKER: Salvatore Cuomo
Full Professor at the University of Naples “Federico II”, Italy
TITLE: Explainable Scientific Machine Learning: Theoretical and Practical Perspectives
ABSTRACT: Scientific Machine Learning (SciML) has revolutionized computational methodologies by integrating model-driven and data-driven approaches. An example is provided by Physics-Informed Neural Networks (PINNs), which have become a deep-learning framework for tackling complex nonlinear partial differential equation (PDE) problems in various scientific fields. This presentation focuses on explainable SciML, addressing several key areas: developing interpretable learning models, integrating domain knowledge, and enhancing computational trust. The talk will discuss strategies for creating transparent neural network models by bridging theoretical insights with practical applications. The discussion will highlight how emerging AI methodologies can advance computational approaches by creating more intelligible, reliable, and adaptable tools that extend the capabilities of traditional numerical methods.
BIO: Salvatore Cuomo is a Full Professor of Numerical Analysis at the Department of Mathematics and Applications “R. Caccioppoli” (DMA) at the University of Naples Federico II (UNINA) in Italy. In 2016, he was a Visiting Researcher at Case Western Reserve University in Ohio, USA, and in 2019, he served as a Visiting Professor at the University of Geosciences in Beijing, China. Professor Cuomo co-founded the Mathematical Modelling and Data Analysis Laboratory (MODAL), the academic spin-off Predico S.r.l. and actually holds the position of Chief Scientific Officer (CSO) at Spici. Additionally, he is a co-inventor of the patent "PredicTS - Prediction Method and Related System." His research interests encompass various topics in Applied Mathematics, specifically: numerical approximation problems (theory, practice, and applications); mathematical aspects of artificial intelligence; and scientific machine learning. He has participated, covering different roles, in numerous research and development projects in scientific computing, data science, the Internet of Things, and Artificial Intelligence. He is also the author of over 190 research papers published in international conferences and journals indexed in the Scopus and WOS databases.
DATE/TIME: Thursday 20 February 2025, at 2.30pm CET.
LOCATION: TBD
The colloquium will also be broadcast online, on Zoom.
Meeting ID: 841 4421 2706 | Passcode: 656302
- Programma
- Ospite
- Participante
- Previsione
- Commenti
I dati meteorologici non sono attualmente disponibili per questa località
Bollettino meteorologico
In corso stec_replace_today_date
stec_replace_current_summary_text
stec_replace_current_temp °stec_replace_current_temp_units
Vento stec_replace_current_wind stec_replace_current_wind_units stec_replace_current_wind_direction
umidità stec_replace_current_humidity %
Sente-se como stec_replace_current_feels_like °stec_replace_current_temp_units
Previsione
Data
Clima
Temperatura
Next 24 Hours
Offerto da openweathermap.org
