Multi-layered Spiking Neural Network with Target Timestamp Threshold Adaptation and STDP
This work addresses the problem of improving energy-efficient SNN performance for machine learning applications, representing an incremental advancement with specific gains.
The paper tackles the performance gap between spiking neural networks (SNNs) and traditional methods by proposing a new threshold adaptation system with timestamp objectives, achieving state-of-the-art classification rates of 98.60% on MNIST and 99.46% on the Faces/Motorbikes dataset using an unsupervised SNN with a linear SVM.
Spiking neural networks (SNNs) are good candidates to produce ultra-energy-efficient hardware. However, the performance of these models is currently behind traditional methods. Introducing multi-layered SNNs is a promising way to reduce this gap. We propose in this paper a new threshold adaptation system which uses a timestamp objective at which neurons should fire. We show that our method leads to state-of-the-art classification rates on the MNIST dataset (98.60%) and the Faces/Motorbikes dataset (99.46%) with an unsupervised SNN followed by a linear SVM. We also investigate the sparsity level of the network by testing different inhibition policies and STDP rules.