Towards a Machine that Works Like the Brain: Neuromorphic Computer

Data manipulation (memory, computation, communications, data mining, sensing) in its many forms drives our modern civilization. The continuous increase in hardware packing density and phenomenal decrease in cost (Moore’s law) has been key to the development of the information revolution. This was fueled by the discovery of revolutionary scientific concepts such as quantum mechanics, coupled with the development of quantum materials and devices. It is however agreed that the enhanced computational capabilities will soon (5-10 years ?) slow down considerably due to a variety of issues which are connected probably to the foundation of the classical Turing-von Neumann paradigm for computing.



At present time new hardware concepts, based on transformative scientific concepts, are needed. This includes reevaluation of data manipulation concepts for software and systems and by necessity will require development of novel hardware including new device and materials concepts. I will describe some of the first steps by a large group of researchers to implement the grand challenge to “develop a machine that works like the brain”.