Aarhus University Seal / Aarhus Universitets segl

Lecture: Advanced Neuromorphic Computing Systems

Postdoc Hooman Farkhani, Department of Engineering, Aarhus University.

22.11.2018 | Jens Kargaard Madsen

Dato tir 27 nov
Tid 13:00 14:00
Sted Room 423, building 5125, Finlandsgade 22, 8200 Aarhus N

Abstract
The brain is a highly complex, but a high performance and ultra-low energy computing system, which benefits from massive parallelism and intertwined network that outperforms the current computers by orders of magnitude, especially when it comes to cognitive computing applications, e.g. pattern recognition. Recently, the reverse engineering of the brain computing, so-called neuromorphic computing, has received much attention thanks to the advances in novel nano-scale technologies. The current implementation of the neuromorphic computing systems (NCS) using Complementary Metal-Oxide-Semiconductor (CMOS) technologies has 5-6 orders of magnitude lower performance (operation/sec/Watt/cm3) compared to the brain. Spintronic devices, using the spin of the electron instead of its charge, have been considered one of the most promising approaches for implementing not only memories but also for NCSs leading to a high density, high speed and energy saving. Although much progress was obtained in the past few years, no real implementation of a high-speed low-energy NCS, filling the gap between the brain and the state-of-the-art computers, was demonstrated. The main goal of my research is to fill the gap between human brain and neuromorphic computing systems.

Forelæsning / foredrag