Locally Connected Spiking Neural Networks for Unsupervised Feature Learning
This work addresses efficient and biologically plausible feature learning in spiking neural networks for image processing, though it appears incremental as it builds on existing SNN and STDP methods.
The paper tackles unsupervised feature learning for image classification using locally connected spiking neural networks (LC-SNNs) with spike-timing-dependent plasticity, achieving state-of-the-art accuracy on two image datasets while requiring fewer parameters and demonstrating robustness to synapse and neuron deletion.
In recent years, Spiking Neural Networks (SNNs) have demonstrated great successes in completing various Machine Learning tasks. We introduce a method for learning image features by \textit{locally connected layers} in SNNs using spike-timing-dependent plasticity (STDP) rule. In our approach, sub-networks compete via competitive inhibitory interactions to learn features from different locations of the input space. These \textit{Locally-Connected SNNs} (LC-SNNs) manifest key topological features of the spatial interaction of biological neurons. We explore biologically inspired n-gram classification approach allowing parallel processing over various patches of the the image space. We report the classification accuracy of simple two-layer LC-SNNs on two image datasets, which match the state-of-art performance and are the first results to date. LC-SNNs have the advantage of fast convergence to a dataset representation, and they require fewer learnable parameters than other SNN approaches with unsupervised learning. Robustness tests demonstrate that LC-SNNs exhibit graceful degradation of performance despite the random deletion of large amounts of synapses and neurons.