Event-Based Control for Online Training of Neural Networks

Abstract

Convolutional Neural Network (CNN) has become the most used method for image classification tasks. During its training the learning rate and the gradient are two key factors to tune for influencing the convergence speed of the model. Usual learning rate strategies are time-based i.e. monotonous decay over time. Recent state-of-the-art techniques focus on adaptive gradient algorithms i.e. Adam and its versions. In this paper we consider an online learning scenario and we propose two Event-Based control loops to adjust the learning rate of a classical algorithm E (Exponential)/PD (Proportional Derivative)-Control. The first Event-Based control loop will be implemented to prevent sudden drop of the learning rate when the model is approaching the optimum. The second Event-Based control loop will decide, based on the learning speed, when to switch to the next data batch. Experimental evaluation is provided using two state-of-the-art machine learning image datasets (CIFAR-10 and CIFAR-100). Results show the Event-Based E/PD is better than the original algorithm (higher final accuracy, lower final loss value), and the Double-Event-Based E/PD can accelerate the training process, save up to 67% training time compared to state-of-the-art algorithms and even result in better performance.

Linear Regression over Networks with Communication Guarantees

“…The technical methodology borrows ideas from the problem of scheduling control tasks over shared communication networks (Eisen et al, 2019;Gatsis et al, 2015;Ayan et al, 2019;Soleymani et al, 2016). The methodology is also related to event-triggered learning that tries to update only if necessary (Solowjow et al, 2018;Zhao et al, 2020).…”