Machine learning methods and their mathematical underpinning
Machine learning (ML) is undergoing a massive expansion, and the last decade has seen a tremendous improvement in the application of ML methods in areas such as (bio-) medical sciences, computer vision and finance, to name a few.
Remarkably, while ML relies on mathematical models and tools, many algorithms, which are used within the field, lack a solid mathematical foundation. Even fundamental questions such as convergence, convergence rates, and the topology and geometry with which data should be studied remain unanswered.
This five-day workshop for researchers working in areas such as approximation theory, inverse problems, optimal transport, multi-scale analysis and statistics will seek to address some of these issues by exploring the mathematical underpinning of ML methods.
The workshop will examine the connection between ML and mathematical disciplines such as numerical analysis, inverse problems, optimisation, statistics, optimal transport, dynamical systems and partial differential equations.
