A Multiplier-Free RNS-Based CNN Accelerator Exploiting Bit-Level Sparsity
- 10 August 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Emerging Topics in Computing
- Vol. 12 (2) , 667-683
Abstract
In this work, a Residue Numbering System (RNS)-based Convolutional Neural Network (CNN) accelerator utilizing a multiplier-free distributed-arithmetic Processing Element (PE) is proposed. A method for maximizing the utilization of the arithmetic hardware resources is presented. It leads to an increase of the system's throughput, by exploiting bit-level sparsity within the weight vectors. The proposed PE design takes advantage of the properties of RNS and Canonical Signed Digit (CSD) encoding to achieve higher energy efficiency and effective processing rate, without requiring any compression mechanism or introducing any approximation. An extensive design space exploration for various parameters (RNS base, PE micro-architecture, encoding) using analytical models as well as experimental results from CNN benchmarks is conducted and the various trade-offs are analyzed. A complete end-to-end RNS accelerator is developed based on the proposed PE. The introduced accelerator is compared to traditional binary and RNS counterparts as well as to other state-of-the-art systems. Implementation results in a 22-nm process show that the proposed PE can lead to 1.85× and 1.54× more energy-efficient processing compared to binary and conventional RNS, respectively, with a 1.88× maximum increase of effective throughput for the employed benchmarks. Compared to a state-of-the-art, all-digital, RNS-based system, the proposed accelerator is 8.87× and 1.11× more energy- and area-efficient, respectively.Keywords
Funding Information
- Khalifa University's SOC Center
- Semiconductor Research Corporation (2020-AH-2983)
This publication has 20 references indexed in Scilit:
- A High-performance RNS LSTM block2022
- On Reducing the Number of Multiplications in RNS-based CNN Accelerators2021
- UCNN: Exploiting Computational Reuse in Deep Neural Networks via Weight Repetition2018
- RNS Hardware Matrix Multiplier for High Precision Neural Network Acceleration: "RNS TPU"2018
- Eyeriss: A Spatial Architecture for Energy-Efficient Dataflow for Convolutional Neural Networks2016
- Deep Residual Learning for Image Recognition2016
- Rethinking the Inception Architecture for Computer Vision2016
- A deep convolutional neural network based on nested residue number system2015
- Efficient canonic signed digit recodingMicroelectronics Journal, 2011
- An O(n) Residue Number System to Mixed Radix Conversion technique2009