A Spike Neural Network Model for Lateral Suppression of Spike-Timing-Dependent Plasticity with Adaptive Threshold
Source abstract
Abstract
Aiming at the practical constraints of high resource occupancy and complex calculations in the existing Spike Neural Network (SNN) image classification model, in order to seek a more lightweight and efficient machine vision solution, this paper proposes an adaptive threshold Spike Neural Network (SNN) model of lateral inhibition of Spike-Timing-Dependent Plasticity (STDP). The conversion from grayscale image to pulse sequence is completed by convolution normalization and first pulse time coding. The network self-classification is realized by combining the classical Spike-Timing-Dependent Plasticity algorithm (STDP) and lateral suppression algorithm. The occurrence of overfitting is effectively suppressed by introducing an adaptive threshold. The experimental results on the MNIST data set show that compared with the traditional SNN classification model, the complexity of the weight update algorithm is reduced from O(n2) to O(1), and the accuracy rate can still remain stable at about 96%. The provided model is conducive to the migration of software algorithms to the bottom layer of the hardware platform, and can provide a reference for the realization of edge computing solutions for small intelligent hardware terminals with high efficiency and low power consumption.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Memory and Neural Computing · CCD and CMOS Imaging Sensors · EEG and Brain-Computer Interfaces
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Electronic Science and Engineering, Nanjing University, Nanjing 210023, China
College of Intelligent Engineering, Nanjing Vocational Institute of Railway Technology, Nanjing 210031, China
Funding
The Natural Science Foundation of the Jiangsu Higher Education Institutions of China
References (30)
Show source-provided references
1. Training a Multi-Layer Photonic Spiking Neural Network with Modified Supervised Learning Algorithm Based on Photonic STDP10.1109/JSTQE.2020.3005589 ↗
2. Quantized STDP-based online-learning spiking neural network10.1007/s00521-021-05832-y ↗
3. Information theory and neural coding10.1038/14731 ↗
4. EEG signal classification using the K-means clustering and a multilayer perceptron neural network model10.1016/j.eswa.2011.04.149 ↗
5. EEG-based spatio–temporal convolutional neural network for driver fatigue evaluation10.1109/TNNLS.2018.2886414 ↗
6. Cascaded LSTM recurrent neural network for automated sleep stage classification using single-channel EEG signals10.1016/j.compbiomed.2019.01.013 ↗
7. Mohammadi, Y., Hajian, M., and Moradi, M.H. (May, January 30). Discrimination of depression levels using machine learning methods on EEG signals. Proceedings of the 27th Iranian Conference on Electrical Engineering (ICEE), Yazd, Iran.
8. NeuCube: A spiking neural network architecture for mapping, learning and understanding of spatio-temporal brain data10.1016/j.neunet.2014.01.006 ↗
9. Hwang, S., Kim, H., and Park, B.G. (2021). Quantized Weight Transfer Method Using Spike-Timing-Dependent Plasticity for Hardware Spiking Neural Network. Appl. Sci., 11.10.3390/app11052059 ↗
10. Digital Hardware Spiking Neuronal Network with STDP for Real-time Pattern Recognition10.2991/jrnal.k.200528.010 ↗
11. A Curiosity-Based Learning Method for Spiking Neural Networks10.3389/fncom.2020.00007 ↗
12. STDP-based spiking deep convolutional neural networks for object recognition10.1016/j.neunet.2017.12.005 ↗
13. Learning generative visual models from few training examples: An incremental Bayesian approach tested on 101 object categories10.1016/j.cviu.2005.09.012 ↗
14. Deep spiking convolutional neural network trained with unsupervised spike-timing-dependent plasticity10.1109/TCDS.2018.2833071 ↗
15. Online supervised learning for hardware-based multilayer spiking neural networks through the modulation of weight-dependent spike-timing-dependent plasticity10.1109/TNNLS.2017.2761335 ↗
16. Gradient-based learning applied to document recognition10.1109/5.726791 ↗
17. First-spike-based visual categorization using reward modulated STDP10.1109/TNNLS.2018.2826721 ↗
18. Unsupervised learning of digit recognition using spike-timing-dependent plasticity10.3389/fncom.2015.00099 ↗
19. Selection and optimization of temporal spike encoding methods for spiking neural networks10.1109/TNNLS.2019.2906158 ↗
20. Neuromorphic sensory systems10.1016/j.conb.2010.03.007 ↗
21. Nuntalid, N., Dhoble, K., and Kasabov, N. (2011, January 13–17). EEG classification with BSA spike encoding algorithm and evolving probabilistic spiking neural network. Proceedings of the 18th International Conference on Neural Information Processing, Shanghai, China.10.1007/978-3-642-24955-6_54 ↗
22. Spike-time encoding as a data compression technique for pattern recognition of temporal data10.1016/j.ins.2017.04.017 ↗
23. Training deep spiking neural networks using backpropagation10.3389/fnins.2016.00508 ↗
24. Computational modeling of spiking neural network with learning rules from STDP and intrinsic plasticity10.1016/j.physa.2017.08.053 ↗
25. Edge computing: Platforms, applications and challenges
26. Loihi: A neuromorphic manycore processor with on-chip learning10.1109/MM.2018.112130359 ↗
27. Efficient FPGA implementations of pair and triplet-based STDP for neuromorphic architectures10.1109/TCSI.2018.2881753 ↗
28. ECG classification algorithm based on STDP and R-STDP neural networks for real-time monitoring on ultra low-power personal wearable devices10.1109/TBCAS.2019.2948920 ↗
29. Supervised learning in spiking neural networks: A review of algorithms and evaluations10.1016/j.neunet.2020.02.011 ↗
30. Indications of nonlinear deterministic and finite dimensional structures in time series of brain electrical activity-dependence on recording region and brain state10.1103/PhysRevE.64.061907 ↗