Sparse Winograd Convolutional neural networks on small-scale systolic arrays

2018/10/03 by Feng Shi, Shi, Feng, Haochen Li +7 · 1 voice
Computer Science · #10010583.10010600.10010628.10010629 #Artificial Intelligence (cs.AI) #Distributed #FOS: Computer and information sciences #Machine Learning (cs.LG) #Parallel #and Cluster Computing (cs.DC) #cs.AI #cs.DC #cs.LG

paper · pdf · doi:10.48550/arxiv.1810.01973

Abstract

The reconfigurability, energy-efficiency, and massive parallelism on FPGAs make them one of the best choices for implementing efficient deep learning accelerators. However, state-of-art implementations seldom consider the balance between high throughput of computation power and the ability of the memory subsystem to support it. In this paper, we implement an accelerator on FPGA by combining the sparse Winograd convolution, clusters of small-scale systolic arrays, and a tailored memory layout design. We also provide an analytical model analysis for the general Winograd convolution algorithm as a design reference. Experimental results on VGG16 show that it achieves very high computational resource utilization, 20x ~ 30x energy efficiency, and more than 5x speedup compared with the dense implementation.

Discussions

• Sparse Winograd CNN on small-scale systolic arrays [hn, 2 points, 0 comments]

Related

• WinoCNN: Kernel Sharing Winograd Systolic Array for Efficient Convolutional Neural Network Acceleration on FPGAs
• Towards Design Space Exploration and Optimization of Fast Algorithms for\n Convolutional Neural Networks (CNNs) on FPGAs
• Packing Sparse Convolutional Neural Networks for Efficient Systolic Array Implementations: Column Combining Under Joint Optimization
• SPOTS: An Accelerator for Sparse Convolutional Networks Leveraging Systolic General Matrix-Matrix Multiplication
• FPGA-based Acceleration for Convolutional Neural Networks: A Comprehensive Review