The talk will explore the evolution of complexity in data sets and the evolution of complexity in the algorithms that deal with them. References will be made to simple datasets demonstrating simple machine learning models up to more complex datasets such as images, text and now non-euclidean datasets - such as point clouds, graphs, manifolds, geospatial datasets and network like data structures that represent highly complex interactions in systems-of-systems. The talk will also make references to a new family of powerful deep learning methods such as Graph neural networks and their applications to non-euclidean data domains.
