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.
Original language | English (US) |
---|---|
Pages (from-to) | 2179-2187 |
Number of pages | 9 |
Journal | Advances in Neural Information Processing Systems |
Volume | 2015-January |
State | Published - 2015 |
Externally published | Yes |
Event | 29th Annual Conference on Neural Information Processing Systems, NIPS 2015 - Montreal, Canada Duration: Dec 7 2015 → Dec 12 2015 |
ASJC Scopus subject areas
- Computer Networks and Communications
- Information Systems
- Signal Processing