Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2014 Jan 30;7:272. doi: 10.3389/fnins.2013.00272

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

Copyright © 2014 Neftci, Das, Pedroni, Kreutz-Delgado and Cauwenberghs.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

PMC Copyright notice

To test recognition accuracy, the trained RBMs are sampled using the I&F neuron-based sampler for up to 1s. The classification is read out by identifying the group of class label neurons that had the highest activity. This experiment is run for RBM parameter sets obtained by standard CD (black, CD) and event-driven CD (green, eCD). To test for robustness to finite precision weights, the RBM was run with parameters obtained by event-driven CD discretized to 8 and 5 bits. In all scenarios, the accuracy after 50ms of sampling was above 80% and after 1s the accuracies typically reached their peak at around 92%. The dashed horizontal lines show the recognition accuracy obtained by minimizing the free-energy (see text). The fact that the eCD curve (solid green) surpasses its free-energy line suggests that a model that is tailored to the I&F spiking neural network was learned.