The Artificial Intelligence (AI) Seminar is a weekly meeting at the University of Alberta where researchers interested in AI can share their research. Presenters include both local speakers from the University of Alberta and visitors from other institutions. Topics related in any way to artificial intelligence, from foundational theoretical work to innovative applications of AI techniques to new fields and problems, are explored.
On May 22, 2020, Canada CIFAR AI Chair at Amii Mark Schmidt, also Associate Professor at the University of British Columbia, Canada Research Chair and Alfred P. Sloan Fellow, presented “Faster Algorithms for Deep Learning?”.
In his talk, Schmidt explains the SGD algorithm, a method which is popular for training deep learning models, but which works slowly due to the variance in the gradient approximation. Schmidt discusses several algorithms that may be implemented to speed up deep learning models, depending on whether the models are under- or over-parameterized.
Watch his full presentation below: