The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here:

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

In-memory computing to solve AI’s energy consumption bottle-neck

Recording from AI Lund lunch seminar 5 May 2021

Ferroelectric memristors for ultra-low power neuromorphic computing.

Title: In-memory computing to solve AI’s energy consumption bottle-neck

When: 5 May at 12.00-13.15

Speaker: Mattias Borg, Dept. of Electrical and Information Technology, LTH , Lund University


The bottle-neck for continued development of Machine Learning lies in the escalating energy consumption during model training. Ultimately, this will require new hardware that implements non-von Neumann architectures, enabling computing-in-memory and even online unsupervised learning by brain-inspired methods.

Memristors based on ferroelectric memory elements are a promising route to such hardware. Here I will introduce memristor-based computing-in-memory, it's benefits in terms of energy-efficiency and our research on the ferroelectric devices that can make it reality.