LightLayers: Parameter Efficient Dense and Convolutional Layers for Image Classification
This work aims to make deep neural network training more accessible for organizations or individuals with limited computational resources, by reducing the number of trainable parameters.
This paper tackles the problem of computationally expensive deep neural network training by proposing LightLayers, a method that reduces the number of trainable parameters in DNNs through matrix factorization. The method achieves promising results on MNIST, Fashion MNIST, and CIFAR-10 datasets, and acceptable performance on CIFAR-100, while using fewer parameters.
Deep Neural Networks (DNNs) have become the de-facto standard in computer vision, as well as in many other pattern recognition tasks. A key drawback of DNNs is that the training phase can be very computationally expensive. Organizations or individuals that cannot afford purchasing state-of-the-art hardware or tapping into cloud-hosted infrastructures may face a long waiting time before the training completes or might not be able to train a model at all. Investigating novel ways to reduce the training time could be a potential solution to alleviate this drawback, and thus enabling more rapid development of new algorithms and models. In this paper, we propose LightLayers, a method for reducing the number of trainable parameters in deep neural networks (DNN). The proposed LightLayers consists of LightDense andLightConv2D layer that are as efficient as regular Conv2D and Dense layers, but uses less parameters. We resort to Matrix Factorization to reduce the complexity of the DNN models resulting into lightweight DNNmodels that require less computational power, without much loss in the accuracy. We have tested LightLayers on MNIST, Fashion MNIST, CI-FAR 10, and CIFAR 100 datasets. Promising results are obtained for MNIST, Fashion MNIST, CIFAR-10 datasets whereas CIFAR 100 shows acceptable performance by using fewer parameters.