Dynamical memristors for higher-complexity neuromorphic computing

doi:10.1038/s41578-022-00434-z

Dynamical memristors for higher-complexity neuromorphic computing

2022-04-08

245

PRE

AI

DOI：10.1038/s41578-022-00434-z

Original

Original request for help

Save

Abstract

En 中文

Research on electronic devices and materials is currently driven by both the slowing down of transistor scaling and the exponential growth of computing needs, which make present digital computing increasingly capacity-limited and power-limited. A promising alternative approach consists in performing computing based on intrinsic device dynamics, such that each device functionally replaces elaborate digital circuits, leading to adaptive 'complex computing'. Memristors are a class of devices that naturally embody higher-order dynamics through their internal electrophysical processes. In this Review, we discuss how novel material properties enable complex dynamics and define different orders of complexity in memristor devices and systems. These native complex dynamics at the device level enable new computing architectures, such as brain-inspired neuromorphic systems, which offer both high energy efficiency and high computing capacity.

Keywords:

NEURAL-NETWORKS

2ND-ORDER MEMRISTOR

MODEL

CHAOS

CLASSIFICATION

ARCHITECTURES

ALGORITHMS

DEVICES

SYSTEMS

VISION

Dynamical memristors for higher-complexity neuromorphic computing

Abstract

Journal

Organization