Random Coordinate Descent Methods for Minimizing Decomposable Submodular Functions
—
filed under:
LFCS Seminar
LFCS Seminar by Dr. Alina Ene (Warwick)
What 


When 
Sep 22, 2015 from 04:00 PM to 05:00 PM 
Where  IF 4.31 
Contact Name  Ilias Diakonikolas 
Add event to calendar 
vCal iCal 
Submodular function minimization is a fundamental optimization problem that arises in several applications in machine learning and computer vision. The problem is known to be solvable in polynomial time, but general purpose algorithms have high running times and are unsuitable for largescale problems. Recent work have used convex optimization techniques to obtain very practical algorithms for minimizing functions that are sums of "simple" functions. In this talk, we give two algorithms based on random coordinate descent with faster linear convergence rates and cheaper iteration costs.
This talk is based on joint work with Huy Nguyen.