Nonlinear Acceleration of CNNs
This work incrementally extends RNA to neural networks, potentially benefiting researchers and practitioners in deep learning optimization.
The paper applied the Regularized Nonlinear Acceleration (RNA) algorithm to optimize CNNs for image recognition on CIFAR10 and ImageNet, resulting in slight improvements in the optimization process after training.
The Regularized Nonlinear Acceleration (RNA) algorithm is an acceleration method capable of improving the rate of convergence of many optimization schemes such as gradient descend, SAGA or SVRG. Until now, its analysis is limited to convex problems, but empirical observations shows that RNA may be extended to wider settings. In this paper, we investigate further the benefits of RNA when applied to neural networks, in particular for the task of image recognition on CIFAR10 and ImageNet. With very few modifications of exiting frameworks, RNA improves slightly the optimization process of CNNs, after training.