In the last 30 years, the theory of Optimal Transportation has emerged as a fertile field of inquiry, and a diverse tool for exploring applications within and beyond mathematics, including Physics, Chemistry, Genomics, Environmental and Earth Sciences, Astronomy and Statistics.

 

More recently, the theory of optimal transport has been pivotal in the context of machine learning research. By using the Optimal Transport distances to guide the update of model parameters, the training process becomes more stable and avoids vanishing gradients. Learning a parametric family of distributions using Optimal Transport geometry enables the model to capture both local and global relationships within the data, making it capable of capturing intricate patterns and dependencies.

 

This thematic program aims to explore mathematical, computational and statistical aspects of Optimal Transport at the interface of Natural Sciences and Statistics. It brings together and fosters collaborations among researchers from complementing mathematical, statistical, quantum physics, quantum chemistry and computational sciences communities that have been or are keen on working on the topic. More specifically, the program focus on:

- Optimal Transport: theory and computational algorithms

- Statistical Optimal Transport

- Wasserstein Gradient Flows in Natural Sciences and Statistics

- Multi-marginal Optimal Transport methods in Electronic Structure Theory

- Trajectory Inference for single-cell RNA and Optimal Transport methods for Omics

- Quantum Optimal Transport\

 

The program also aims to develop new analytical, geometrical, computational, and statistical tools to tackle fundamental issues and to extend applicability to a wider range of problems in Electronic Structure Theory, Quantum Information, Statistics and Computational Omics.