Ruoshi Li
ABSTRACT
Spiking neural networks (SNNs) obtain impressive good performance on various applications due to their powerful computing capacity for encoding spatio-temporal information. However, most existing spiking neural networks remain shallow structures, lacking effective structures to handle real-world tasks. In this work, we propose an autoencoder induced deep spiking neural network (AE-DSN) to improve the representative capacity. Specifically, AE-DSN consists of three coding modules shared the same structure. Each module contains an autoencoder and a spiking coding layer. In particular, we present a progressive training strategy to train these modules one-by-one. For one module, the spiking coding layer is trained using ReSuMe algorithm, guided by the encoded information from the autoencoder, which would be dropped when training the subsequent module. The entire training process terminates when the final spiking coding layer is trained well. Experimental results show that the proposed AE-DSN could effectively extract discriminative features for the input images to achieve superior classification performance.
- Alireza Bagheri, Osvaldo Simeone, and Bipin Rajendran. 2018. Training probabilistic spiking neural networks with first-to-spike decoding. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2986–2990.Google Scholar
Digital Library
- Chih-Chung Chang and Chih-Jen Lin. 2011. LIBSVM: a library for support vector machines. ACM transactions on intelligent systems and technology (TIST) 2, 3 (2011), 1–27.Google ScholarDigital Library
- André Grüning and Ioana Sporea. 2012. Supervised learning of logical operations in layered spiking neural networks with spike train encoding.Neural Processing Letters 36, 2 (2012), 117–134.Google Scholar
- Shiqi Guo and Tong Lin. 2021. An Efficient non-Backpropagation Method for Training Spiking Neural Networks. In 2021 IEEE 33rd International Conference on Tools with Artificial Intelligence (ICTAI). IEEE, 192–199.Google ScholarCross Ref
- Donald Olding Hebb. 2005. The organization of behavior: A neuropsychological theory. Psychology Press.Google Scholar
- Xiaoyu Hu and Chongxin Liu. 2019. Dynamic property analysis and circuit implementation of simplified memristive Hodgkin–Huxley neuron model. Nonlinear Dynamics 97, 2 (2019), 1721–1733.Google ScholarCross Ref
- Laxmi R Iyer and Yansong Chua. 2020. Classifying neuromorphic datasets with tempotron and spike timing dependent plasticity. In 2020 International Joint Conference on Neural Networks (IJCNN). IEEE, 1–8.Google ScholarCross Ref
- Amirhossein Tavanaei and Anthony Maida. 2019. BP-STDP: Approximating backpropagation using spike timing dependent plasticity. Neurocom- puting 330 (2019), 39–47.Google ScholarCross Ref
- John J Wade, Liam J McDaid, Jose A Santos, and Heather M Sayers. 2010. SWAT: a spiking neural network training algorithm for classification problems. IEEE transactions on neural networks 21, 11 (2010), 1817–1830.Google Scholar
- Xiangwen Wang, Xianghong Lin, and Xiaochao Dang. 2020. Supervised learning in spiking neural networks: A review of algorithms and evaluations. Neural Networks 125 (2020), 258–280.Google ScholarCross Ref
- Fu Xing, Ye Yuan, Hong Huo, and Tao Fang. 2019. Homeostasis-based cnn-to-snn conversion of inception and residual architectures. In International Conference on Neural Information Processing. Springer, 173–184.Google ScholarDigital Library
- Qiang Yu, Huajin Tang, Kay Chen Tan, and Haoyong Yu. 2014. A brain-inspired spiking neural network model with temporal encoding and learning. Neurocomputing 138 (2014), 3–13.Google ScholarCross Ref
- Tielin Zhang, Yi Zeng, Dongcheng Zhao, and Mengting Shi. 2018. A plasticity-centric approach to train the non-differential spiking neural networks. In Thirty-second AAAI conference on artificial intelligence.Google ScholarCross Ref
- Junhong Zhao, Jacek M Zurada, Jie Yang, and Wei Wu. 2018. The convergence analysis of SpikeProp algorithm with smoothing L1/ 2 regularization. Neural Networks 103 (2018), 19–28.Google ScholarDigital Library
Index Terms
- Autoencoder Induced Deep Spiking Neural Network
Recommendations
Deep learning in spiking neural networks
AbstractIn recent years, deep learning has revolutionized the field of machine learning, for computer vision in particular. In this approach, a deep (multilayer) artificial neural network (ANN) is trained, most often in a supervised manner using ...
A Deep-Learning-Based Method For Spike Sorting With Spiking Neural Network
ICBSP '22: Proceedings of the 2022 7th International Conference on Biomedical Imaging, Signal ProcessingSpike sorting is a technique of detecting and assigning signals generated by the neurons of the brain and classifying which spike belongs to which neurons. Spike sorting is usually done by a clustering algorithm that groups the spikes coming from the ...
Comments