20th Presentation | A Probabilistic Machine Learning Framework for Evaluating Energy-Dependent Fission Product Yields | Vaia Prassa

20th Presentation | A Probabilistic Machine Learning Framework for Evaluating Energy-Dependent Fission Product Yields | Vaia Prassa

NanoAI welcomes Vaia Prassa  who will present his work, titled “A Probabilistic Machine Learning Framework for Evaluating Energy-Dependent Fission Product Yields”.

As always, an open discussion will follow.

When? November 7th, 15:00 Greece

Description:

Fission product yields (FPYs) are essential for advancing nuclear science and technology, particularly in areas involving nuclear structure and reaction dynamics. However, traditional computational methods and theoretical models often face limitations due to inherent constraints and a lack of experimental data, posing challenges in obtaining accurate and comprehensive fission data.

In this talk, I will present a probabilistic machine learning approach that combines Mixture Density Networks (MDNs) and Gaussian Process Regression (GPR) to evaluate energy-dependent fission yields. MDNs are employed to learn from existing data, predict unknown yields, and simultaneously quantify uncertainties, making them especially effective in scenarios with incomplete experimental information. GPR is utilized to capture the distributions of single-fission yields, generating high-quality synthetic samples that serve as valuable inputs for the MDNs.

The MDN framework demonstrates reliable accuracy in capturing both the distribution shapes and energy dependencies of FPYs. This approach offers a promising solution to the challenges of traditional methods, providing a robust tool for FPY evaluation even in the absence of complete experimental data.

Dr. Vaia Prassa is an Assistant Professor at the Department of Physics, University of Thessaly, with expertise in nuclear theory, quantum physics, and reaction dynamics, focusing on computational modeling and simulations. She completed her PhD in Theoretical Physics and her MSc in Computational Physics i at Aristotle University of Thessaloniki.

Her career includes multiple postdoctoral positions across Europe at institutions such as the University of Thessaly, University of Zagreb, University of Jyväskylä, and Aristotle University. Dr. Prassa has received notable fellowships from the Stavros Niarchos Foundation, Marie Skłodowska Curie COFUND-NEWFELPRO, Finland Distinguished Professor Programme (FiDiPro), and the Greek Scholarship Foundation.

Dr. Prassa has published 42 peer-reviewed articles and has been an invited speaker at prominent international conferences and seminars, including EURISOL 2013, KITPC in Beijing, and the Institute for Nuclear Theory at the University of Washington. She has also given talks at renowned institutions, such as GANIL in France, the National Center for Scientific Research (NCSR) Democritus and the National and Kapodistrian University of Athens.

Speaker links:

Subscribe to our mailing list: https://lists.iit.demokritos.gr/mailman/listinfo/nanoai
You can find more info on this upcoming talk as well as on our previous talks at http://nano-ai.iit.demokritos.gr/