Jun Yang

Project title

Geometry-Aware Monte Carlo Sampling (GAMeS)

What is your project about?

The GAMeS project develops Monte Carlo methods for high-dimensional models, large datasets, and parallel computing. By using geometric information in model spaces, it designs efficient algorithms for inference in complex systems. Building on recent breakthroughs in heavy-tailed sampling and interacting particle systems, GAMeS advances scalable and reliable methods for data science and AI.

How did you become interested in your particular field of research?

I became interested in Monte Carlo methods through a fascination with how randomness can solve complex problems. During my studies, I was drawn to the interplay between probability, geometry, and computation in Markov chain Monte Carlo (MCMC). This led me to explore both the theory and practice of MCMC, focusing on making it scalable and reliable for modern challenges in data science and AI.

What are the scientific challenges and perspectives in your project?

The main challenges lie in designing Monte Carlo algorithms that remain efficient in high dimensions, scale to large datasets, and exploit parallel computing. GAMeS addresses these by leveraging geometric insights and recent advances in particle methods. The project aims to create theoretically grounded, practical tools that advance scalable inference and open new perspectives in statistical computing and AI.

What is your estimate of the impact, which your project may have to society in the long term?

GAMeS aims to improve the reliability and scalability of algorithms used in data science and AI, which are increasingly central to society. In the long term, this research can support better decision-making in areas like healthcare, climate modeling, and public policy, by enabling trustworthy inference from complex data and promoting transparency in AI-driven systems.

Which impact do you expect the Sapere Aude programme will have on your career as a researcher?

Being part of the Sapere Aude program is a major milestone in my research career. It provides the opportunity to build an independent research group, strengthen international collaborations, and pursue ambitious ideas at the frontier of statistical computing. The program will accelerate my development as a research leader and help shape the future of reliable, scalable methods in data science and AI.