Abstract

Machine learning has experienced unprecedented growth in recent years, often referred to as an “artificial intelligence revolution.” Biological systems inspire the fundamental approach for this new computing paradigm: using neural networks to classify large amounts of data into sorting categories. Current machine learning schemes implement simulated neurons and synapses on standard computers based on a von Neumann architecture. This approach is inefficient in energy consumption, and thermal management, motivating searching for hardware-based systems that imitate the brain. This tutorial describes the present state of thermal management of neuromorphic computing technology and describes the challenges and opportunities of the energy-efficient implementation of neuromorphic devices. The introduction is presented in chapter 1, where we briefly describe the main features of brain-inspired computing and quantum materials for implementing neuromorphic devices. In chapter 2 we discuss the brain criticality and resistive switching-based neuromorphic devices. Chapter 3 presents the energy and electrical considerations for spiking-based computation. We address the fundamental features of the brain's thermal regulation in chapter 4. Hereafter, we analyse the physical mechanisms for thermal management (chapter 5) and thermoelectric control of materials and neuromorphic devices (chapter 6). At the end we describe challenges and new avenues for implementing energy-efficient computing.