Abstract
Neuromorphic hardware inspired by the brain has attracted much attention for its advanced information processing concept. However, implementing online learning in the neuromorphic chip is still challenging. In this paper, we present a bio-plausible online-learning spiking neural network (SNN) model for hardware implementation. The SNN consists of an input layer, an excitatory layer, and an inhibitory layer. To save resource cost and accelerate information processing speed during hardware implementation, online learning based on the spiking neural model is realized by trace-based spiking-timing-dependent plasticity (STDP). Neuron and synapse activities are digitalized, and decay behaviors of neuron and synapse parameters are realized by the bit-shift operation. After learning training set from the Modified National Institute of Standards and Technology (MNIST), the spiking neural model successfully recognizes the digits from the MNIST test set, showing the feasibility and capability of the model. The recognition accuracy increases significantly from 90.0% to 94.5% with the number of the excitatory/inhibitory neurons rising from 400 to 3,500, which provides a guide to make a trade-off between the recognition accuracy and the resource cost during hardware implementation. Encouragingly, compared to its corresponding floating-point model, the proposed model reduces the hardware resources and power consumption by 40.7% and 36.3%, respectively (under 55-nm CMOS process).
Highlights
A neuromorphic computing platform, inspired by the advanced information processing scheme of the brain [1], is more efficient and bio-plausible than the traditional Von Neumann computing platform when dealing with brain-like computation tasks [2]
LEARNING PROCESS The spiking neural network (SNN) will be used to classify digital images after it learns from 60,000 images of the Modified National Institute of Standards and Technology (MNIST) training set
The weights gradually gathered to 0, which is similar to the sparse connectivity concept in the deep neural networks and is bio-plausible for each neuron only connects to a limited number of neurons in the biological neural networks [40]
Summary
A neuromorphic computing platform, inspired by the advanced information processing scheme of the brain [1], is more efficient and bio-plausible than the traditional Von Neumann computing platform when dealing with brain-like computation tasks (such as pattern recognition) [2]. To further optimize the hardware implementation with the bit-shift operation (see below), pre is reset to an initial value (e.g., 104) when the pre-synaptic neuron fires a spike, and a learning rate is introduced in the weight modification: wij (t) = wij (t − 1) − λpre × postj(t − 1). It is assumed that the pointer points to W(0) at the current time step and one of the pre-synaptic neurons connected to this neuron fires a spike. The weight of this synapse (w0) and the delay time (assumed to be 4.5 ms) are retrieved from the weight matrix Mxe and the delay matrix (as shown in Fig. 1), respectively. The above procedures are repeated until the end of the learning or inferencing process
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
