Francisco Criado Gallart
01.01.2021 − 31.12.2022
Worst-case complexity bounds are increasingly insufficient to explain the (often superior) real-world performance of optimization and learning algorithms. We will consider data-dependent rates, approximation guarantees, and complexity bounds to provide guarantees much more in line with actual performance. We are in particular interested in exploiting properties of the feasible region to obtain algorithms that are adaptive to the problem structure.
A more technical overview can be found [here].