View source for JuNe - The Juelich Network

=== Overview ===
Sensory cortices undergo a massive bombardment of input. Network dynamics upon thalamocortical stimulation shows a stereotyped spread of activation which can be largely explained by the connection probabilities. The interplay between stimulation, feed-forward and feedback, as well as recurrent connections causes asynchronous irregular activity and oscillations. This model aims at a massively parallel simulation of interconnected local cortical networks, each comprising in the order of 80,000 neurons. Ideally, the model should be operational on the wafer-scale system as well as on the Jugene cluster.

A description of this project can also be found here: https://brainscales.kip.uni-heidelberg.de/internal/wiki/index.php/WP_5_index

== Developing institutions ==
FZJ

== Developers ==
Current developers are S. van Albada, M. Diesmann and S. Schrader

== Results of the ESS workshop, Oct 2011 ==

The model was presented at the ''Using the ESS + Neuromorphic hardware'' workshop in Dresden, Oct 4th, 2011 by Sacha van Albada. The slides of the presentation can be found [https://brainscales.kip.uni-heidelberg.de/internal/wiki/images/f/fb/JuNe.pdf here]

* An initial model of JuNe has been implemented in PyNN and is operational on the ESS.
* The original neuron model has been transferred to the hardware neuron (AdEx).
* The necessity of massive Poisson input has to be discussed in the presence of hardware limitations. Several strategies to ease bandwidth have been proposed and implemented in PyNN.
* Synaptic homogeneities (delays, weights) may surpass hardware limitations, but the former can possibly be approximated with distributed time constants, if necessary.