We present first experimental results on the novel BrainScaleS-2 neuromorphic architecture based on an analog neuro-synaptic core and augmented by embedded microprocessors for complex plasticity and experiment control. The high acceleration factor of 1000 compared to biological dynamics enables the execution of computationally expensive tasks, by allowing the fast emulation of long-duration experiments or rapid iteration over many consecutive trials. The flexibility of our architecture is demonstrated in a suite of five distinct experiments, which emphasize different aspects of the BrainScaleS-2 system.
@article{arxiv.1912.12980,
title = {Versatile emulation of spiking neural networks on an accelerated neuromorphic substrate},
author = {Sebastian Billaudelle and Yannik Stradmann and Korbinian Schreiber and Benjamin Cramer and Andreas Baumbach and Dominik Dold and Julian Göltz and Akos F. Kungl and Timo C. Wunderlich and Andreas Hartel and Eric Müller and Oliver Breitwieser and Christian Mauch and Mitja Kleider and Andreas Grübl and David Stöckel and Christian Pehle and Arthur Heimbrecht and Philipp Spilger and Gerd Kiene and Vitali Karasenko and Walter Senn and Mihai A. Petrovici and Johannes Schemmel and Karlheinz Meier},
journal= {arXiv preprint arXiv:1912.12980},
year = {2022}
}