articleFeb 2, 2017GREEN OA

FINN

YUYaman UmurogluNJNicholas J. FraserGGGiulio GambardellaMBMichaela BlottPLPhilip Leong

Norwegian University of Science and Technology · Xilinx (Ireland) · +2 more institutions

Indexed inarxivcrossref

Abstract

Research has shown that convolutional neural networks contain significant redundancy, and high classification accuracy can be obtained even when weights and activations are reduced from floating point to binary values. In this paper, we present FINN, a framework for building fast and flexible FPGA accelerators using a flexible heterogeneous streaming architecture. By utilizing a novel set of optimizations that enable efficient mapping of binarized neural networks to hardware, we implement fully connected, convolutional and pooling layers, with per-layer compute resources being tailored to user-provided throughput requirements. On a ZC706 embedded FPGA platform drawing less than 25 W total system power, we…

Citation impact

995
total citations
FWCI
30.31
Percentile
100%
References
27
Citations per year

Authors

7
  • YU
    Yaman UmurogluCorresponding

    Norwegian University of Science and Technology

  • NJ
    Nicholas J. Fraser

    Xilinx (Ireland)

  • GG
    Giulio Gambardella

    Xilinx (Ireland)

  • MB
    Michaela Blott

    Xilinx (Ireland)

  • PL
    Philip Leong

    University of Sydney

Topics & keywords

Topics
Keywords
  • MNIST database
  • Pooling
  • Convolutional neural network
  • Field-programmable gate array
  • Latency (audio)
  • Binary number
  • Contextual image classification
  • Pattern recognition (psychology)
No related works found for this paper.

Funding