An asynchronous parallel stochastic coordinate descent algorithm for minimizing smooth unconstrained or separably constrained functions. The method achieves a linear convergence rate on functions that satisfy an essential strong convexity property and a sublinear rate (1=K) on general convex functions. Nearlinear speedup on a multicore system can be expected if the number of processors is O(n1=2) in unconstrained optimization and O(n1=4) in the separable-constrained case, where n is the number of variables. … Asynchronous Parallel Stochastic Coordinate Descent (ASYCD)
If you did not already know: “Asynchronous Parallel Stochastic Coordinate Descent (ASYCD)”
08 Monday Jun 2015
Posted What is ...
in