Variational Stochastic Gradient Descent for Deep Neural Networks

arXiv:2404.06549v1 Announce Type: new
Abstract: Optimizing deep neural networks is one of the main tasks in successful deep learning. Current state-of-the-art optimizers are adaptive gradient-based optimization methods such as Adam. Recently, there has been an increasing interest in formulating gradient-based optimizers in a probabilistic framework for better estimation of gradients and modeling uncertainties. Here, we propose to combine both approaches, resulting in the Variational Stochastic Gradient Descent (VSGD) optimizer. We model gradient updates as a probabilistic model and utilize stochastic …

abstract adam art arxiv cs.lg current deep learning framework gradient modeling networks neural networks optimization state stat.ml stochastic tasks type