Return to Colloquia & Seminar listing
"SGD and Randomized Projections methods for linear systems"
PDE & Applied Mathematics| Speaker: | Prof. Deanna Needell, Claremont McKenna College |
| Location: | 2112 MSB |
| Start time: | Thu, Dec 11 2014, 3:10PM |
Description
In this talk we will give a brief overview of stochastic
gradient pursuit and the closely related Kaczmarz method for solving
linear systems, or more generally convex optimization problems. We
will present some new results which tie these methods together and
prove the best known convergence rates for these methods under mild
Lipschitz conditions. The methods empirically and theoretically rely
on probability distributions to dictate the order of sampling in the
algorithms. It turns out that the choice of distribution may
drastically change the performance of the algorithm, and the theory
has only begun to explain this phenomenon.
Talk moved to 3:10pm.