The unprecedented development of Artificial Intelligence and the Internet of Things has led to smart edge devices that can sense and think, such as smartphones, smartwatches, and smart glasses. Inspired by the human nervous system, running computation and perception entirely in memory has been proposed as the most promising approach to create machinery with autonomous intelligence. Breakthroughs in new materials and nanodevices offer unprecedented potential for incorporating sensing, memory, and computing in an event-based and parallel manner much like biological organisms. Such in-memory computing and sensing naturally resolve the energy and time bottlenecks incurred by sequential digitization of analog sensory signals and frequent data shuttling between processing and memory units, which largely benefit the edge smart devices in terms of throughput and energy efficiency. These could revolutionize the computing and sensing paradigm and would greatly propel a broad spectrum of applications such as healthcare informatics, autonomous vehicles, and business analytics. In this Research Topic, we focus our efforts to assemble state-of-the-art reports on new materials, nanodevices, and computational algorithms that cater to the area of neuro-inspired computing and sensing.
The exponential growth of edge sensory nodes, driven by the Internet of Things and Big Data, leads to a vast amount of data generated at the edge every second, incurring large energy and time overhead to transmit the sensory data to the cloud for central processing. On the other hand, the human brain and the associated peripheral nervous system outperform most modern-day electronic processors and sensors in their computational and data transmission capability, and power efficiency, thanks to the capability to process and store analog signals right at where they are generated. The quest to match this cognitive proficiency at ultralow power is the driving force for the field of neuromorphic computing and sensing.
This Research Topic aims to provide a comprehensive overview of emerging materials, devices, circuits and algorithms that enable in-memory computing and neuro-inspired event-based sensing protocols, including but not limited to memristive materials, devices, circuit topologies and neural network prototypes. We also foresee emerging opportunities fusing neuro-inspired sensing technologies, wearable and flexible/stretchable electronics, and novel neuromorphic platforms, and seek an interdisciplinary perspective that integrates materials science, electronic engineering, computer science, and neuroscience. The selected articles will present a state-of-the-art overview of the progress in this topic area.
We invite the submission of Original Research, Review, Mini Review, Perspective articles on themes including, but not limited to:
• Rational design or selection of materials and devices that portray computing-in-memory and/or sensing-in-memory capabilities
• Understanding in-memory computing/sensing devices, including both experimental investigation and theoretical formulation
• Novel circuit and system designs for computing-in-memory and/or sensing-in-memory
• Algorithms and applications enabled by computing-in-memory and/or sensing-in-memory systems.
The unprecedented development of Artificial Intelligence and the Internet of Things has led to smart edge devices that can sense and think, such as smartphones, smartwatches, and smart glasses. Inspired by the human nervous system, running computation and perception entirely in memory has been proposed as the most promising approach to create machinery with autonomous intelligence. Breakthroughs in new materials and nanodevices offer unprecedented potential for incorporating sensing, memory, and computing in an event-based and parallel manner much like biological organisms. Such in-memory computing and sensing naturally resolve the energy and time bottlenecks incurred by sequential digitization of analog sensory signals and frequent data shuttling between processing and memory units, which largely benefit the edge smart devices in terms of throughput and energy efficiency. These could revolutionize the computing and sensing paradigm and would greatly propel a broad spectrum of applications such as healthcare informatics, autonomous vehicles, and business analytics. In this Research Topic, we focus our efforts to assemble state-of-the-art reports on new materials, nanodevices, and computational algorithms that cater to the area of neuro-inspired computing and sensing.
The exponential growth of edge sensory nodes, driven by the Internet of Things and Big Data, leads to a vast amount of data generated at the edge every second, incurring large energy and time overhead to transmit the sensory data to the cloud for central processing. On the other hand, the human brain and the associated peripheral nervous system outperform most modern-day electronic processors and sensors in their computational and data transmission capability, and power efficiency, thanks to the capability to process and store analog signals right at where they are generated. The quest to match this cognitive proficiency at ultralow power is the driving force for the field of neuromorphic computing and sensing.
This Research Topic aims to provide a comprehensive overview of emerging materials, devices, circuits and algorithms that enable in-memory computing and neuro-inspired event-based sensing protocols, including but not limited to memristive materials, devices, circuit topologies and neural network prototypes. We also foresee emerging opportunities fusing neuro-inspired sensing technologies, wearable and flexible/stretchable electronics, and novel neuromorphic platforms, and seek an interdisciplinary perspective that integrates materials science, electronic engineering, computer science, and neuroscience. The selected articles will present a state-of-the-art overview of the progress in this topic area.
We invite the submission of Original Research, Review, Mini Review, Perspective articles on themes including, but not limited to:
• Rational design or selection of materials and devices that portray computing-in-memory and/or sensing-in-memory capabilities
• Understanding in-memory computing/sensing devices, including both experimental investigation and theoretical formulation
• Novel circuit and system designs for computing-in-memory and/or sensing-in-memory
• Algorithms and applications enabled by computing-in-memory and/or sensing-in-memory systems.
