Optimizing Convolutional Neural Networks for Embedded Systems by Means of Neuroevolution
This work addresses the need for energy-efficient neural networks in resource-constrained embedded systems, representing an incremental improvement in automated CNN design.
The paper tackles the problem of designing efficient convolutional neural networks for embedded systems by proposing a neuroevolution method that optimizes both classification error and model complexity, achieving competitive results on MNIST and CIFAR-10 benchmarks.
Automated design methods for convolutional neural networks (CNNs) have recently been developed in order to increase the design productivity. We propose a neuroevolution method capable of evolving and optimizing CNNs with respect to the classification error and CNN complexity (expressed as the number of tunable CNN parameters), in which the inference phase can partly be executed using fixed point operations to further reduce power consumption. Experimental results are obtained with TinyDNN framework and presented using two common image classification benchmark problems -- MNIST and CIFAR-10.