This repository contains the implementation of the Ternary Spiking Neural Network (TSNN) architecture. The TSNN architecture offers significant model compression, achieving up to 16x compression compared to traditional neural networks. Moreover, it achieves impressive accuracy results on various datasets: 98.43% on N-MNIST, 89.07% on CIFAR-10, and 65.24% on CIFAR-100, using 4 time steps.
If you find this work useful, please consider citing the original paper:
Wu, M., Kan, Y., Zhang, R., & Nakashima, Y. (2022, September). GAND-Nets: Training Deep Spiking Neural Networks with Ternary Weights. In 2022 IEEE 35th International System-on-Chip Conference (SOCC) (pp. 1-6). IEEE.
For any questions or issues, please feel free to reach out to us via email at [email protected].