A Spiking Neural Network Model of the Lateral Geniculate Nucleus on the SpiNNaker Machine

Abstract

We present a spiking neural network model of the thalamic Lateral Geniculate Nucleus (LGN) developed on SpiNNaker, which is a state-of-the-art digital neuromorphic hardware built with very-low-power ARM processors. The parallel, event-based data processing in SpiNNaker makes it viable for building massively parallel neuro-computational frameworks. The LGN model has 140 neurons representing a “basic building block” for larger modular architectures. The motivation of this work is to simulate biologically plausible LGN dynamics on SpiNNaker. Synaptic layout of the model is consistent with biology. The model response is validated with existing literature reporting entrainment in steady state visually evoked potentials (SSVEP)—brain oscillations corresponding to periodic visual stimuli recorded via electroencephalography (EEG). Periodic stimulus to the model is provided by: a synthetic spike-train with inter-spike-intervals in the range 10–50 Hz at a resolution of 1 Hz; and spike-train output from a state-of-the-art electronic retina subjected to a light emitting diode flashing at 10, 20, and 40 Hz, simulating real-world visual stimulus to the model. The resolution of simulation is 0.1 ms to ensure solution accuracy for the underlying differential equations defining Izhikevichs neuron model. Under this constraint, 1 s of model simulation time is executed in 10 s real time on SpiNNaker; this is because simulations on SpiNNaker work in real time for time-steps dt ⩾ 1 ms. The model output shows entrainment with both sets of input and contains harmonic components of the fundamental frequency. However, suppressing the feed-forward inhibition in the circuit produces subharmonics within the gamma band (>30 Hz) implying a reduced information transmission fidelity. These model predictions agree with recent lumped-parameter computational model-based predictions, using conventional computers. Scalability of the framework is demonstrated by a multi-node architecture consisting of three “nodes,” where each node is the “basic building block” LGN model. This 420 neuron model is tested with synthetic periodic stimulus at 10 Hz to all the nodes. The model output is the average of the outputs from all nodes, and conforms to the above-mentioned predictions of each node. Power consumption for model simulation on SpiNNaker is ≪1 W.

Foveal-pit inspired filtering of DVS spike response

“…We have used a 128 × 128 sized neuromorphic DVS developed by Serrano-Gotarredona et al [13] to capture nonsynthetic visual data. Each pixel of the DVS processes the input continuously and emits a spike based on the variation in the illumination impinging upon it [19]. A sample illustration is provided in Fig.…”

In Search of Trustworthy and Transparent Intelligent Systems With Human-Like Cognitive and Reasoning Capabilities

“…Consequently, if an image recognition system exploits these concepts at the feature extraction stage, it would resemble more a biological system and is likely to yield better results. There have been several attempts to develop computational models of LGN neurons (Einevoll and Halnes, 2015 ; Sen-Bhattacharya et al, 2017 ). We need to investigate how these models can be adapted to develop machine learning systems for computer vision.…”

sPyNNaker: A Software Package for Running PyNN Simulations on SpiNNaker

“…For example, this flexibility has led to the SpiNNaker platform executing large-scale networks of neurons (Sen-Bhattacharya et al, 2017 ; van Albada et al, 2018 ), and performing analysis of cognitive tasks such as action selection (Sen-Bhattacharya et al, 2018 ). Applications of learning in SNNs have also been investigated, such as the study of learning based on Bayesian inference in Knight et al ( 2016 ), and reinforcement learning in Mikaitis et al ( 2018 ).…”