Supervised Learning in Multilayer Spiking Neural Networks With Spike Temporal Error Backpropagation

The brain-inspired spiking neural networks (SNNs) hold the advantages of lower power consumption and powerful computing capability. However, the lack of effective learning algorithms has obstructed the theoretical advance and applications of SNNs. The majority of the existing learning algorithms for SNNs are based on the synaptic weight adjustment. However, neuroscience findings confirm that synaptic delays can also be modulated to play an important role in the learning process. Here, we propose a gradient descent-based learning algorithm for synaptic delays to enhance the sequential learning performance of single spiking neuron. Moreover, we extend the proposed method to multilayer SNNs with spike temporal-based error backpropagation. In the proposed multilayer learning algorithm, information is encoded in the relative timing of individual neuronal spikes, and learning is performed based on the exact derivatives of the postsynaptic spike times with respect to presynaptic spike times. Experimental results on both synthetic and realistic datasets show significant improvements in learning efficiency and accuracy over the existing spike temporal-based learning algorithms. We also evaluate the proposed learning method in an SNN-based multimodal computational model for audiovisual pattern recognition, and it achieves better performance compared with its counterparts.

I. Introduction

The great success of deep artificial neural network (ANN) in image recognition [1], [2], object detection [3], [4], speech recognition [5], and other application fields has attracted a large number of researchers to follow up. However, due to its high demand for computing resources (e.g., GPUs and computing clusters), ANN’s application in power-constrained platforms, such as edge computing, is greatly limited [6]. Spiking neural network (SNN) is proposed to simulate the information processing in the brain, and the spike-based manner makes it suitable for ultralow-power, event-driven neuromorphic implementation [7]. In addition, it imitates the dynamic process of biological neurons in more detail, and this makes it possible to capture more types of information, such as timing, phase, and oscillation. By virtue of its strong computing capability and low power consumption, SNN becomes the potential direction of the future development in artificial intelligence.