On the Almost Sure Convergence of Stochastic Gradient Descent in Non-Convex Problems
Panayotis Mertikopoulos,Nadav Hallak,Ali Kavis,Volkan Cevher
In this paper, we analyze the trajectories of stochastic gradient descent (SGD) with the aim of understanding their convergence properties in non-convex problems. We first show that the sequence of iterates generated by SGD remains bounded and converges with probability $1$ under a very broad range of step-size schedules. Subsequently, we prove that the algorithms rate of convergence to local minimizers with a positive-definite Hessian is $O(1/n^p)$ if the method is run with a $u0398(1/n^p)$ step-size. This provides an important guideline for tuning the algorithms step-size as it suggests that a cool-down phase with a vanishing step-size could lead to significant performance gains; we demonstrate this heuristic using ResNet architectures on CIFAR. Finally, going beyond existing positive probability guarantees, we show that SGD avoids strict saddle points/manifolds with probability $1$ for the entire spectrum of step-size policies considered.
