About this Research Topic
Traditional computing systems based on the von Neumann architecture are facing severe problems related to memory-access bottleneck and energy efficiency wall. Indeed, the amount of data to be processed is exploding day by day in the Internet-of-Things era and the continuous scaling of devices becomes harder year by year. To address these problems, new computing hardware is being based on extreme-parallel architecture, inspired by synaptic plasticity in the brain, which is capable of in-memory computing and is suitable for multi-valued or analog arithmetic.
Memristors can be useful for realizing new computing hardware satisfying the conditions mentioned above. They are non-volatile memory devices that are fast and energy-efficient during read and write operations. Memristors are fabricated in a CMOS-compatible process and can compute analog arithmetic. These features make memory-based computing architecture promising for the realization of neuromorphic systems. In the future, such systems could solve the problems of memory access bottlenecks and energy efficiency walls.
This Research Topic aims to collect the most advanced results of research and developments on memristor-based neuromorphic computing. More specifically, this topic’s interests include the use of memristor devices, circuits, systems, applications, algorithms, etc., for the implementation of neuromorphic systems. in which CMOS and memristors can be integrated together to process vast amounts of unstructured data from various Internet-of-Things sensors. Furthermore, the aim of this collection is to summarize and review the recent important contributions to memristor-computing-based neuromorphic techniques, in order to predict future memristive hardware of non-Von-Neumann computing.
Relevant topics include (but are not limited to):
-Theoretical advances in memristor-computing for realizing neuromorphic systems
-Memristor-computing materials and devices
-Memristor-CMOS hybrid circuits and systems for realizing neuromorphic hardware
-Neuromorphic applications, learning algorithms for memristor-computing-based AI hardware
The resulting collection of original research articles, reviews, and commentaries will be a reference for the research on memristor-computing-based neuromorphic systems, advancing the research further through discussions and new collaborations in our community.
Memristors can be useful for realizing new computing hardware satisfying the conditions mentioned above. They are non-volatile memory devices that are fast and energy-efficient during read and write operations. Memristors are fabricated in a CMOS-compatible process and can compute analog arithmetic. These features make memory-based computing architecture promising for the realization of neuromorphic systems. In the future, such systems could solve the problems of memory access bottlenecks and energy efficiency walls.
This Research Topic aims to collect the most advanced results of research and developments on memristor-based neuromorphic computing. More specifically, this topic’s interests include the use of memristor devices, circuits, systems, applications, algorithms, etc., for the implementation of neuromorphic systems. in which CMOS and memristors can be integrated together to process vast amounts of unstructured data from various Internet-of-Things sensors. Furthermore, the aim of this collection is to summarize and review the recent important contributions to memristor-computing-based neuromorphic techniques, in order to predict future memristive hardware of non-Von-Neumann computing.
Relevant topics include (but are not limited to):
-Theoretical advances in memristor-computing for realizing neuromorphic systems
-Memristor-computing materials and devices
-Memristor-CMOS hybrid circuits and systems for realizing neuromorphic hardware
-Neuromorphic applications, learning algorithms for memristor-computing-based AI hardware
The resulting collection of original research articles, reviews, and commentaries will be a reference for the research on memristor-computing-based neuromorphic systems, advancing the research further through discussions and new collaborations in our community.
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.
