Affiliation:
1. Intel Corporation, USA
Abstract
Neuromorphic computing is a rapidly emerging field that seeks to emulate the computational principles of the brain using novel materials and devices. While traditional computing architectures, such as the von Neumann architecture, have experienced exponential improvements in computational power due to the continuous shrinkage of transistor technology, the laws of physics will eventually limit the performance of these systems. Neuromorphic computing offers the potential to continue scaling technology by incorporating novel functional materials and devices into new computing architectures. In this chapter, the authors will explore the principles of neuromorphic computing and the key challenges that must be overcome in order to fully implement these systems. The authors will discuss the role of new nanostructured materials and devices in neuromorphic computing, and the importance of understanding charge and defects migration, fault tolerance, and manufacturability.