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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Spiking Neural Networks (SNNs) offer a novel computational paradigm thatcaptures some of the efficiency of biological brains by processing throughbinary neural dynamic activations. Probabilistic SNN models are typicallytrained to maximize the likelihood of the desired outputs by using unbiasedestimates of the log-likelihood gradients. While prior work used single-sampleestimators obtained from a single run of the network, this paper proposes toleverage multiple compartments that sample independent spiking signals whilesharing synaptic weights. The key idea is to use these signals to obtain moreaccurate statistical estimates of the log-likelihood training criterion, aswell as of its gradient. The approach is based on generalizedexpectation-maximization (GEM), which optimizes a tighter approximation of thelog-likelihood using importance sampling. The derived online learning algorithmimplements a three-factor rule with global per-compartment learning signals.Experimental results on a classification task on the neuromorphic MNIST-DVSdata set demonstrate significant improvements in terms of log-likelihood,accuracy, and calibration when increasing the number of compartments used fortraining and inference.

Multi-Sample Online Learning for Spiking Neural Networks based on Generalized Expectation Maximization