Stochastic convex optimization is a basic and well studied primitive in machine learning. It is well known that convex and Lipschitz functions can be minimized efficiently using Stochastic Gradient Descent (SGD). The Normalized Gradient Descent (NGD) algorithm, is an adaptation of Gradient Descent, which updates according to the direction of the gradients, rather than the gradients themselves. In this paper we analyze a stochastic version of NGD and prove its convergence to a global minimum for a wider class of functions: we require the functions to be quasi-convex and locally-Lipschitz. Quasi-convexity broadens the concept of unimodality to multidimensions and allows for certain types of saddle points, which are a known hurdle for first-order optimization methods such as gradient descent. Locally-Lipschitz functions are only required to be Lipschitz in a small region around the optimum. This assumption circumvents gradient explosion, which is another known hurdle for gradient descent variants. Interestingly, unlike the vanilla SGD algorithm, the stochastic normalized gradient descent algorithm provably requires a minimal minibatch size.
|Original language||American English|
|Number of pages||9|
|Journal||Advances in Neural Information Processing Systems|
|State||Published - 2015|
|Event||29th Annual Conference on Neural Information Processing Systems, NIPS 2015 - Montreal, Canada|
Duration: 7 Dec 2015 → 12 Dec 2015
Bibliographical noteFunding Information:
The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreement no 336078-ERC-SUBLRN. Shai S-Shwartz is supported by ISF no 1673/14 and by Intel's ICRI-CI.