What is it about?
Neuromorphic computers are brain-inspired architectures, based on massively parallel interconnected neurons that communicate via spikes. They potentially can reduce energy usage of computation by orders of magnitude, but we currently don't understand yet what sort of problems they can or cannot do efficiently. This paper offers a mathematical framework for investigating this question from a theoretical point of view.
Featured Image
Photo by Hal Gatewood on Unsplash
Why is it important?
Reducing carbon footprint of computation is pivotal giving the ever increasing computational demands of ICT and AI systems. If we can reduce energy usage of computation significantly this will help mitigate the climate crisis.
Read the Original
This page is a summary of: On the computational power and complexity of Spiking Neural Networks, March 2020, ACM (Association for Computing Machinery),
DOI: 10.1145/3381755.3381760.
You can read the full text:
Contributors
The following have contributed to this page
