HOME
News
Accepted: "Constructive community race: full-density spiking neural network model drives neuromorphic computing"
We implemented a wide-spread neuromorphic benchmark model to evaluate BrainScaleS hardware performance for a use case from computational neuroscience. Using a downscaled variant of the full model, we generated metrics showcasing order-of-magnitude efficiency improvements on the BrainScaleS-1 analog substrate compared to conventional computing methods, including large-scale FPGA substrates.
It will be published in IOP Neuromorphic Computing and Engineering; a preprint is available on arXiv.
Last update of this page: 2026-01-13 by Electronic Visions
