Timezone: »

 
Poster
Local Smoothness in Variance Reduced Optimization
Daniel Vainsencher · Han Liu · Tong Zhang

Wed Dec 09 04:00 PM -- 08:59 PM (PST) @ 210 C #72 #None

Abstract We propose a family of non-uniform sampling strategies to provably speed up a class of stochastic optimization algorithms with linear convergence including Stochastic Variance Reduced Gradient (SVRG) and Stochastic Dual Coordinate Ascent (SDCA). For a large family of penalized empirical risk minimization problems, our methods exploit data dependent local smoothness of the loss functions near the optimum, while maintaining convergence guarantees. Our bounds are the first to quantify the advantage gained from local smoothness which are significant for some problems significantly better. Empirically, we provide thorough numerical results to back up our theory. Additionally we present algorithms exploiting local smoothness in more aggressive ways, which perform even better in practice.

Author Information

Daniel Vainsencher (Princeton University)
Han Liu (Princeton University)
Tong Zhang (Rutgers)

More from the Same Authors