Corpus ID: 232168654

unzipFPGA: Enhancing FPGA-based CNN Engines with On-the-Fly Weights Generation

@article{Venieris2021unzipFPGAEF,
  title={unzipFPGA: Enhancing FPGA-based CNN Engines with On-the-Fly Weights Generation},
  author={Stylianos I. Venieris and J. Fern{\'a}ndez-Marqu{\'e}s and N. Lane},
  journal={ArXiv},
  year={2021},
  volume={abs/2103.05600}
}
Single computation engines have become a popular design choice for FPGA-based convolutional neural networks (CNNs) enabling the deployment of diverse models without fabric reconfiguration. This flexibility, however, often comes with significantly reduced performance on memory-bound layers and resource underutilisation due to suboptimal mapping of certain layers on the engine’s fixed configuration. In this work, we investigate the implications in terms of CNN engine design for a class of models… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 89 REFERENCES
Toolflows for Mapping Convolutional Neural Networks on FPGAs
  • 92
  • PDF
DNNVM: End-to-End Compiler Leveraging Heterogeneous Optimizations on FPGA-Based CNN Accelerators
  • Yu Xing, S. Liang, +6 authors Yi Shan
  • Computer Science
  • IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
  • 2020
  • 20
  • Highly Influential
  • PDF
f-CNNx: A Toolflow for Mapping Multiple Convolutional Neural Networks on FPGAs
  • 8
  • PDF
Automatic Generation of Multi-Precision Multi-Arithmetic CNN Accelerators for FPGAs
  • Y. Zhao, Xitong Gao, +6 authors C. Xu
  • Computer Science, Engineering
  • 2019 International Conference on Field-Programmable Technology (ICFPT)
  • 2019
  • 7
  • PDF
Angel-Eye: A Complete Design Flow for Mapping CNN Onto Embedded FPGA
  • K. Guo, Lingzhi Sui, +6 authors H. Yang
  • Computer Science
  • IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
  • 2018
  • 188
  • PDF
An OpenCL(TM) Deep Learning Accelerator on Arria 10
  • 78
Latency-driven design for FPGA-based convolutional neural networks
  • 52
  • PDF
Light-OPU: An FPGA-based Overlay Processor for Lightweight Convolutional Neural Networks
  • 5
  • Highly Influential
A high performance FPGA-based accelerator for large-scale convolutional neural networks
  • 174
  • PDF
DEF: Differential Encoding of Featuremaps for Low Power Convolutional Neural Network Accelerators
  • 1
...
1
2
3
4
5
...