About this Research Topic
Memristor (MR) device is a promising candidate for emerging applications inspired by neurobiological forms. This is due to the unique properties of MR devices with the ability to have variable levels of resistance and non-volatility. The programming of MR resistance in response to input stimuli can be directly related to altered synaptic plasticity in the brain, opening new opportunities for bio-inspired and bio-memetic electronics that are not possible using traditional electronics technologies. The MR device has a history-dependent property that is not found in silicon-based devices. This gives MR technology the privilege to play a role in going beyond von Neumann architecture for neuromorphic and AI applications, through enabling In-Memory-Computing (IMC) for both digital- and analog-type processing, where both computation and storage operations are occurring within the same device. Such an approach is very favorable in energy-constraint systems such as IoT and autonomous systems, where both local processing and decision-making are critical.
The goal of this Research Topic is to enrich the state of the art of neuromorphic engineering through the design, fabrication, and characterization of novel MR devices suitable for neuromorphic applications. Building on the unique and competitive features of MR devices including their low power, high density, and high-speed operations, novel neuromorphic architectures could be developed to overcome the limitations associated with current silicon-based integrated circuit technologies. This topic targets high-impact contributions that provide an evidence-based translation that demonstrates the impact and potential of MR devices and their associated properties on the field of neuromorphic engineering.
The Research Topic will focus on various aspects related to the deployment of MR devices for neuromorphic and AI applications. This will include the design, fabrication, and characterization of novel MR stacks that can provide competitive behavior to mimic the synaptic plasticity of neuronal behavior in the brain. High priority will be given to papers that report novel nanoscale MR synapses with linear flux-charge relationships to allow full analog switching and their deployment in neuromorphic computing architectures. MR devices that exhibit adaptive synapse-like conductivity in response to non-voltage input stimuli are of great interest, too. This Research Topic will also encourage papers that exploit novel MR device/crossbar that has the ability to compute, process, and maintain information in parallel, within the same element.
The goal of this Research Topic is to enrich the state of the art of neuromorphic engineering through the design, fabrication, and characterization of novel MR devices suitable for neuromorphic applications. Building on the unique and competitive features of MR devices including their low power, high density, and high-speed operations, novel neuromorphic architectures could be developed to overcome the limitations associated with current silicon-based integrated circuit technologies. This topic targets high-impact contributions that provide an evidence-based translation that demonstrates the impact and potential of MR devices and their associated properties on the field of neuromorphic engineering.
The Research Topic will focus on various aspects related to the deployment of MR devices for neuromorphic and AI applications. This will include the design, fabrication, and characterization of novel MR stacks that can provide competitive behavior to mimic the synaptic plasticity of neuronal behavior in the brain. High priority will be given to papers that report novel nanoscale MR synapses with linear flux-charge relationships to allow full analog switching and their deployment in neuromorphic computing architectures. MR devices that exhibit adaptive synapse-like conductivity in response to non-voltage input stimuli are of great interest, too. This Research Topic will also encourage papers that exploit novel MR device/crossbar that has the ability to compute, process, and maintain information in parallel, within the same element.
Keywords: Memristor Crossbar, In-memory-computing, Synaptic electronics, Deep learning, AI Hardware, Memristor Fabrication, Smart Architecture, Memristor modelling, Accelerators, Artificial Neural Networks, Spiking Neural Networks
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
