Excited to share that our latest work “A CMOS-compatible oscillation-based VO₂ Ising machine solver” has been published in Nature Communications.
In this article, we demonstrate how to solve complex optimization problems with a neural network made of vanadium dioxide (VO₂) oscillators. The neuromorphic architecture is scalable, power-efficient, and capable of processing data locally. The technology is ideal for specialized hardware accelerators in AI applications and edge computing.

We are proud that it is even highlighted on the dedicated Editors’Highlight page on ‘devices‘, which showcases the 50 best papers recently published in an area.

Congratulations to our PhD candidate Olivier Maher from IBM Research and his team for this outstanding work!