English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Spiking Neural Networks (SNNs) with a large number of weights and variedweight distribution can be difficult to implement in emerging in-memorycomputing hardware due to the limitations on crossbar size (implementing dotproduct), the constrained number of conductance levels in non-CMOS devices andthe power budget. We present a sparse SNN topology where non-criticalconnections are pruned to reduce the network size and the remaining criticalsynapses are weight quantized to accommodate for limited conductance levels.Pruning is based on the power law weight-dependent Spike Timing DependentPlasticity (STDP) model; synapses between pre- and post-neuron with high spikecorrelation are retained, whereas synapses with low correlation or uncorrelatedspiking activity are pruned. The weights of the retained connections arequantized to the available number of conductance levels. The process of pruningnon-critical connections and quantizing the weights of critical synapses isperformed at regular intervals during training. We evaluated our sparse andquantized network on MNIST dataset and on a subset of images from Caltech-101dataset. The compressed topology achieved a classification accuracy of 90.1%(91.6%) on the MNIST (Caltech-101) dataset with 3.1x (2.2x) and 4x (2.6x)improvement in energy and area, respectively. The compressed topology is energyand area efficient while maintaining the same classification accuracy of a2-layer fully connected SNN topology.

STDP Based Pruning of Connections and Weight Quantization in Spiking Neural Networks for Energy Efficient Recognition