We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.AR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Hardware Architecture

Title: BinArray: A Scalable Hardware Accelerator for Binary Approximated CNNs

Authors: Mario Fischer, Juergen Wassner (Department of Engineering and Architecture, Lucerne University of Applied Sciences and Arts, Switzerland)
Abstract: Deep Convolutional Neural Networks (CNNs) have become state-of-the art for computer vision and other signal processing tasks due to their superior accuracy. In recent years, large efforts have been made to reduce the computational costs of CNNs in order to achieve real-time operation on low-power embedded devices.
Towards this goal we present BinArray, a custom hardware accelerator for CNNs with binary approximated weights. The binary approximation used in this paper is an improved version of a network compression technique initially suggested in [1]. It drastically reduces the number of multiplications required per inference with no or very little accuracy degradation. BinArray easily scales and allows to compromise between hardware resource usage and throughput by means of three design parameters transparent to the user. Furthermore, it is possible to select between high accuracy or throughput dynamically during runtime. BinArray has been optimized at the register transfer level and operates at 400 MHz as instruction-set processor within a heterogenous XC7Z045-2 FPGA-SoC platform.
Experimental results show that BinArray scales to match the performance of other accelerators like EdgeTPU [2] for different network sizes. Even for the largest MobileNet only 50% of the target device and only 96 DSP blocks are utilized.
Comments: 11 pages, 11 figures
Subjects: Hardware Architecture (cs.AR); Signal Processing (eess.SP)
DOI: 10.1109/CCWC51732.2021.9375963
Cite as: arXiv:2012.03481 [cs.AR]
  (or arXiv:2012.03481v1 [cs.AR] for this version)

Submission history

From: Juergen Wassner [view email]
[v1] Mon, 7 Dec 2020 07:02:40 GMT (1354kb,D)

Link back to: arXiv, form interface, contact.