An Efficient Accelerator for Multiple Convolutions From the Sparsity Perspective
- 4 March 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. PP (10638210) , 1-5
Abstract
Convolutional neural networks (CNNs) have emerged as one of the most popular ways applied in many fields. These networks deliver better performance when going deeper and larger. However, the complicated computation and huge storage impede hardware implementation. To address the problem, quantized networks are proposed. Besides, various convolutional structures are designed to meet the requirements of different applications. For example, compared with the traditional convolutions (CONVs) for image classification, CONVs for image generation are usually composed of traditional CONVs, dilated CONVs, and transposed CONVs, leading to a difficult hardware mapping problem. In this brief, we translate the difficult mapping problem into the sparsity problem and propose an efficient hardware architecture for sparse binary and ternary CNNs by exploiting the sparsity and low bit-width characteristics. To this end, we propose an ineffectual data removing (IDR) mechanism to remove both the regular and irregular sparsity based on dual-channel processing elements (PEs). Besides, a flexible layered load balance (LLB) mechanism is introduced to alleviate the load imbalance. The accelerator is implemented with 65-nm technology with a core size of 2.56 mm^2. It can achieve 3.72-TOPS/W energy efficiency at 50.1 mW, which makes it a promising design for embedded devices.Keywords
This publication has 11 references indexed in Scilit:
- DT-CNN: Dilated and Transposed Convolution Neural Network Accelerator for Real-Time Image Segmentation on Mobile Devices2019
- Enhancing Utilization of SIMD-Like Accelerator for Sparse Convolutional Neural NetworksIEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2019
- An Energy-Efficient Reconfigurable Processor for Binary-and Ternary-Weight Neural Networks With Flexible Data Bit WidthIEEE Journal of Solid-State Circuits, 2018
- PIM-TGAN: A Processing-in-Memory Accelerator for Ternary Generative Adversarial Networks2018
- GNA: Reconfigurable and Efficient Architecture for Generative Network AccelerationIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2018
- Towards Efficient Microarchitectural Design for Accelerating Unsupervised GAN-Based Deep Learning2018
- SCNN2017
- Energy-Efficient Design of Processing Element for Convolutional Neural NetworkIEEE Transactions on Circuits and Systems II: Express Briefs, 2017
- YodaNN: An Architecture for Ultralow Power Binary-Weight CNN AccelerationIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2017
- Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural NetworksIEEE Journal of Solid-State Circuits, 2016