Toward the Optimal Design and FPGA Implementation of Spiking Neural Networks
Type
ArticleKAUST Department
Electrical Engineering ProgramElectrical Engineering
Physical Science and Engineering (PSE) Division
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Date
2021Submitted Date
2020-02-18Permanent link to this record
http://hdl.handle.net/10754/667399
Metadata
Show full item recordAbstract
The performance of a biologically plausible spiking neural network (SNN) largely depends on the model parameters and neural dynamics. This article proposes a parameter optimization scheme for improving the performance of a biologically plausible SNN and a parallel on-field-programmable gate array (FPGA) online learning neuromorphic platform for the digital implementation based on two numerical methods, namely, the Euler and third-order Runge-Kutta (RK3) methods. The optimization scheme explores the impact of biological time constants on information transmission in the SNN and improves the convergence rate of the SNN on digit recognition with a suitable choice of the time constants. The parallel digital implementation leads to a significant speedup over software simulation on a general-purpose CPU. The parallel implementation with the Euler method enables around 180x (20x) training (inference) speedup over a Pytorch-based SNN simulation on CPU. Moreover, compared with previous work, our parallel implementation shows more than 300x (240x) improvement on speed and 180x (250x) reduction in energy consumption for training (inference). In addition, due to the high-order accuracy, the RK3 method is demonstrated to gain 2x training speedup over the Euler method, which makes it suitable for online training in real-time applications.Citation
Guo, W., Yantir, H. E., Fouda, M. E., Eltawil, A. M., & Salama, K. N. (2021). Toward the Optimal Design and FPGA Implementation of Spiking Neural Networks. IEEE Transactions on Neural Networks and Learning Systems, 1–15. doi:10.1109/tnnls.2021.3055421Sponsors
This work was supported by the King Abdullah University of Science and Technology (KAUST) AI Initiative, Saudi ArabiaPublisher
IEEEAdditional Links
https://ieeexplore.ieee.org/document/9353400/https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9353400
ae974a485f413a2113503eed53cd6c53
10.1109/TNNLS.2021.3055421