A DSP-Based Precision-Scalable MAC With Hybrid Dataflow for Arbitrary-Basis-Quantization CNN Accelerator
- 24 February 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 24 (1) , 65-68
Abstract
Precision-scalable convolutional neural networks (CNNs) offer a promising solution to balance network accuracy and hardware efficiency, facilitating high-performance execution on embedded devices. However, the requirement for small fine-grained multiplication calculations in precision-scalable (PS) networks has resulted in limited exploration on FPGA platforms. It is found that the deployment of PS accelerators encounters the following challenges: LUT-based multiply-accumulates (MACs) fail to make full use of DSP, and DSP-based MACs support limited precision combinations and cannot efficiently utilize DSP. Therefore, this brief proposes a DSP-based precision-scalable MAC with hybrid dataflow that supports most precision combinations and ensures high-efficiency utilization of DSP and LUT resources. Evaluating on mixed 4b/8b VGG16, compared with 8b baseline, the proposed accelerator achieves 3.97× improvement in performance with only a 0.37% accuracy degradation. Additionally, compared with state-of-the-art accelerators, the proposed accelerator achieves 1.20×-2.69× improvement in DSP efficiency and 1.63×-6.34× improvement in LUT efficiency.Keywords
Funding Information
- Science and Technology Planning Project of Guangzhou (2023B01J0007)
- National Natural Science Foundation of China (62301165)
- Basic and Applied Basic Research Foundation of Guangdong (2022A1515110774)
This publication has 12 references indexed in Scilit:
- A Flexible FPGA-Based Accelerator for Efficient Inference of Multi-Precision CNNs2024
- ALSCA: A Large-Scale Sparse CNN Accelerator Using Position-First Dataflow and Input Channel Merging ApproachIEEE Transactions on Circuits and Systems II: Express Briefs, 2024
- A Tiny Accelerator for Mixed-Bit Sparse CNN Based on Efficient Fetch Method of SIMO SPadIEEE Transactions on Circuits and Systems II: Express Briefs, 2023
- Mixed-Precision Neural Network Quantization via Learned Layer-Wise Importance2022
- A High Performance Multi-Bit-Width Booth Vector Systolic Accelerator for NAS Optimized Deep Learning Neural NetworksIEEE Transactions on Circuits and Systems I: Regular Papers, 2022
- High Performance CNN Accelerators Based on Hardware and Algorithm Co-OptimizationIEEE Transactions on Circuits and Systems I: Regular Papers, 2020
- LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networks2018
- Loom: Exploiting Weight and Activation Precisions to Accelerate Convolutional Neural Networks2018
- Bit-width Adaptive Accelerator Design for Convolution Neural Network2018
- Evaluating Fast Algorithms for Convolutional Neural Networks on FPGAs2017