Optimal transport (OT) provides a powerful and flexible way to compare probability measures, of all shapes: absolutely continuous, degenerate, or discrete. This includes of course point clouds, histograms of features, and more generally datasets, parametric densities or generative models. Originally proposed by Monge in the eighteenth century, this theory later led to Nobel Prizes for Koopmans and Kantorovich as well as Villani's and Figalli's Fields Medals in 2010 and 2018.
After having attracted the interest of mathematicians for several years, OT has recently reached the machine learning community, because it can now tackle (both in theory and numerically) challenging learning scenarios, including for instance dimensionality reduction and structured prediction problems that involve histograms or point clouds, and estimation of parametric densities or generative models in highly degenerate / high-dimensional problems.
Following its first edition in 2014 and second edition in 2017, this workshop will be a renewed opportunity to present recent progress and improved understanding in this developing field. We are in particular interested in discussing new methodological, theoretical and computational achievements and insights, as well as highlighting new practical applications of OT for machine learning and related fields.
The format of the workshop will encourage discussion and presentation of contributed content in addition to our lineup of invited talks.
Long talks: (tentative)
Short talks: