This is a talk given by Prof Silvia Villa as part of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization, https://congreso.us.es/mlneedsmo/
First order optimization methods are the algorithms of choice for modern machine earning and inverse problems. In this talk, I will show that the trajectories generated by the chosen algorithm, if suitably initialized, converges to some properly defined minimal “norm” solution. I will discuss the advantages of such approaches with respect to explicit (Tykhonov) regularization strategies and some numerical examples.
Link to the talk: https://eu.bbcollab.com/collab/ui/session/guest/dd1cc465c8ee463e97bc8eaf177dda0f
The organizers of the Online Seminar Series Machine Learning NeEDS Mathematical Optimization
Emilio Carrizosa, IMUS-Instituto de Matemáticas de la Universidad de Sevilla
Thomas Halskov, Copenhagen Business School
Kseniia Kurishchenko, Copenhagen Business School
Cristina Molero-Río, IMUS-Instituto de Matemáticas de la Universidad de Sevilla
Jasone Ramírez-Ayerbe, IMUS-Instituto de Matemáticas de la Universidad de Sevilla
Dolores Romero Morales, Copenhagen Business School