Figure 5 | Scientific Reports

Figure 5

From: Stochastic binary synapses having sigmoidal cumulative distribution functions for unsupervised learning with spike timing-dependent plasticity

Figure 5

Effect of variability on memory maintenance and recognition accuracy. (a) Memory maintenance characteristics obtained with sigmoidal stochastic S-STDP in the case of \(k=4\) (with \(p=0.2\) and \(q=0.08\)), with \(\sigma\) being varied from 0 to 2.0 as well as that obtained with deterministic S-STDP as a reference (taken from Fig. 1b). (b) Recognition accuracy after training with 150,000 samples as a function of \(\sigma\) for \(k=3\) and \(4\).

Back to article page