About the webinar
Scientific Machine Learning is a rapidly evolving field that combines machine learning, artificial intelligence, and traditional scientific computing. Its application in Neuroscience is at the forefront of this field, bridging the gap between classical computational modeling and state-of-the-art AI. These applications range from replacing traditional partial differential equation solvers with neural surrogates to evaluating the computational complexity of single neurons. In this talk, we will examine some of these methodologies, highlighting their inherent strengths and limitations, as well as the emerging pathways being defined within this growing field.
Who is the webinar for?
Master’s and PhD students, researchers, professors, and anyone with a basic understanding of AI and machine learning.
Key takeaways for participants:
-
Understanding the strengths and weaknesses of neural surrogates in Computational Neuroscience.
-
Identifying emerging pathways in the rapidly growing field of Scientific Machine Learning applied to Computational Neuroscience.
Speaker bio:
Luca Pellegrini is a PhD student in the Joint PhD Program in Computational Mathematics and Decision Sciences at the University of Pavia (UniPv) and the University of Lugano (USI). His research focuses on applying neural networks to computational electrophysiology. In particular, he focuses on exploring Scientific Machine Learning methods, such as neural operators and physics-informed neural networks, to solve stiff ionic problems. He also works on reproducing the input-output mapping of Purkinje cells through causality-respecting networks. Additionally, he investigates hybrid methods that combine neural networks with classical numerical solvers to leverage the strengths of neural networks with classical numerical solvers.
Linkedin: www.linkedin.com/in/pellegrini-luca