M2DA: A Low-Complex Design Methodology for Convolutional Neural Network Exploiting Data Symmetry and Redundancy
- PDF / 2,110,781 Bytes
- 26 Pages / 439.37 x 666.142 pts Page_size
- 6 Downloads / 185 Views
M2DA: A Low-Complex Design Methodology for Convolutional Neural Network Exploiting Data Symmetry and Redundancy Madhuri Panwar1 · Nemani Sri Hari2 · Dwaipayan Biswas3 · Amit Acharyya1 Received: 14 January 2020 / Revised: 19 August 2020 / Accepted: 23 August 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Convolutional neural network (CNN) is one of the most dominant deep learning networks with good generalization ability. Its high performance in solving large and complex learning problems has enabled usability in IoT devices. However, CNN involves a substantial amount of convolution operations, which demand a large number of power-consuming multipliers. This hinders the deployment of deep CNNs on mobile and IoT edge devices owing to restricted power–area constraints. In this paper, we propose a low-complex methodology named ‘minimal modified distributed arithmetic’ (M2DA) for convolutional neural network (CNN) by exploiting the data symmetry and consequently storing only the unique kernel coefficient’s combinations and the size of required memory and multiplication operations can be reduced, leading to power–area efficient design. For validation, a low-complex CNN architecture for activity recognition application is designed and synthesized in Synopsys using the UMC 65 nm technology wherein average 36.89% and 51.63% improvement is achieved in power and area, respectively, compared to conventional MDA methodology. To demonstrate the significance of the proposed M2DA methodology, we have also implemented the Alexnet which is the most widely and publicly available CNN model for the image classification problem. Keywords Deep learning · Convolutional neural network · Modified distributed arithmetic (MDA) · Multiplier-less design
1 Introduction Recent breakthroughs in the development of convolutional neural networks (CNNs) have revolutionized a variety of fields including but not limited to computer vision [17],
B
Amit Acharyya [email protected]
Extended author information available on the last page of the article
Circuits, Systems, and Signal Processing
artificial intelligence [12], speech recognition [11], activity monitoring, image related data analytics [18] and language processing [9]. It uses multiple layers to extract the high-level features to classify complex problems, outperforming conventional machine learning models [43]. The strong generalization ability of this multilayered network has pushed the classification to the human accuracy, which has enabled applications in Internet-of-Things (IoT)-based devices. These IoT devices are usually battery powered or rely on energy harvester, having limited energy sources owing to low cost and portability issues [8, 10, 28]. On the other hand, CNN requires a substantial amount of computations, which become a critical issue for the deployment of CNN in the IoT devices. The main compute-intensive operations in CNN are performed by convolutional and fully connected layers which involve a substantial amount of multiplicatio
Data Loading...
