Nonlinear Acceleration of Stochastic Algorithms

Abstract

Extrapolation methods use the last few iterates of an optimization algorithm to produce a better estimate of the optimum. They were shown to achieve optimal convergence rates in a deterministic setting using simple gradient iterates. Here, we study extrapolation methods in a stochastic setting, where the iterates are produced by either a simple or an accelerated stochastic gradient algorithm. We first derive convergence bounds for arbitrary, potentially biased perturbations, then produce asymptotic bounds using the ratio between the variance of the noise and the accuracy of the current point. Finally, we apply this acceleration technique to stochastic algorithms such as SGD, SAGA, SVRG and Katyusha in different settings, and show significant performance gains.

Cited by in corpus (6)

• Globally Convergent Type-I Anderson Acceleration for Non-Smooth Fixed-Point Iterations
• A Generic Acceleration Framework for Stochastic Composite Optimization
• Geometry of First-Order Methods and Adaptive Acceleration
• Stochastic Anderson Mixing for Nonconvex Stochastic Optimization
• Nonlinear Collaborative Scheme for Deep Neural Networks
• Generalization of Quasi-Newton Methods: Application to Robust Symmetric Multisecant Updates