Reconstructing Spiking Neural Networks Using a Single Neuron with Autapses

Researchers propose TDA-SNN, a novel spiking neural network framework that uses a single neuron with time-delayed autapses to reconstruct traditional multilayer architectures. The approach significantly reduces neuron count and memory requirements while maintaining competitive performance, though at the cost of increased temporal latency.

• →TDA-SNN framework reconstructs complex spiking neural networks using just a single leaky integrate-and-fire neuron with autapses.
• →The approach can emulate reservoir, multilayer perceptron, and convolution-like architectures within a unified framework.
• →Experiments show competitive performance on sequential, event-based, and image benchmarks compared to standard SNNs.
• →The method greatly reduces neuron count and state memory requirements while increasing per-neuron information capacity.
• →Trade-off exists between space efficiency and temporal latency in extreme single-neuron configurations.