Future AI Hardware - Emerging technologies bring Neuromorphic computing closer to the Brain

No ratings

Presented at InfoSecurityDenmark 2020 by

Mr. Moradi

The brain is a highly complex, high performance and low energy computing system due to its massive parallelism and intertwined network, which outperforms the current computers by orders of magnitudes, especially for cognitive computing applications. Although, years of research have been spent on studying the brain for clinical purposes, now, engineers and technologists are interested to understand how the brain works in such a low-energy manner too. In past few years, many large interdisciplinary projects have been kicked-off to achieve a deeper understanding of how the brain processes the data; An understanding that will help not only the clinical setup to treat neurological disorders, but also engineers to mimic the brain's behavior on computing.A large effort has been made into understanding the computing and mimicking the brain into an artificial implementation, so-calledneuromorphic computing. 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. Therefore, the main need is to explore novel technologies enabling the development of brain-like neuromorphic computing systems; not just mimicking the brain, but learning how the brain works and applying it to build novel computing system.In this talk, we will focus on novel technologies enabling the development of such...