About this Research Topic
The brain is capable of solving complicated tasks, such as one-shot learning, abstraction and sensory-motor control, with an extremely low power consumption of few tens of Watts. That is made possible by (i) the large connectivity among the computing elements inside it (every neuron is, on average, connected to about 10,000 other neurons), (ii) the spike-based communication/processing protocol adopted (energy is consumed only when/where it is required), (iii) the co-localization of memory and computing functionalities, and (iv) the high specialization of its building blocks, such as somas, synapses and dendrites. Brain-inspired hardware aims at mimicking the successful structure and operation of the brain to perform the variety of complex tasks for which the latter appears to be the most effective and high-performance solution. In such hardware, realistic artificial counterparts for the biological actors playing a role in brain structure and operation are needed. Conceiving and developing those counterparts represents an ambitious goal, requiring the investigation and the exploitation of novel materials, devices, and solutions.
This Research Topic aims at summarizing the most recent advances in the field of artificial elements for brain-inspired hardware, with a focus on sensing and computing applications. Papers will provide a comprehensive picture of the variety of innovations that are today investigated and explored to make artificial components mimicking the biological building blocks of the brain involved in sensing and computing. This picture will serve as a starting point for the design of complex neuromorphic systems targeting artificial intelligence and high-performance machine learning.
The scope of the research works presented in this Research Topic should be the reproduction of the neural phenomena involved in biological sensing (e.g., direction selectivity in the retina, echolocation) and computing (neuron integrate/fire, synaptic plasticity, pattern learning and recognition) through advanced materials, novel solid-state devices or innovative operating schemes for mainstream and emerging solid-state components. Special attention should be devoted to the physical effects and properties exploited to achieve the neuromorphic operation of the investigated artificial elements.
Topics of interest include, but are not limited to:
- advanced materials or material properties for neuromorphic hardware for sensing and computing
- novel solid-state devices for neuromorphics hardware for sensing and computing
- new physical effects in materials and solid-state devices exploitable for neuromorphic hardware for sensing and computing
- innovative operating schemes for mainstream and emerging solid-state devices allowing to reproduce principles and functionalities needed in neuromorphic hardware for sensing and computing
Papers targeting this Research Topic should provide a minimum experimental validation of the reported results (pure theoretical or simulation-only works are not suitable for the Research Topic). Research works on hardware accelerators for artificial neural networks and in-memory computing are out of the scope of the Research Topic if not closely linked to brain-inspired subjects. Research works on novel materials and devices for weight storage that do not mimic biological details are out of the scope of this Research Topic as well.
This Research Topic aims at summarizing the most recent advances in the field of artificial elements for brain-inspired hardware, with a focus on sensing and computing applications. Papers will provide a comprehensive picture of the variety of innovations that are today investigated and explored to make artificial components mimicking the biological building blocks of the brain involved in sensing and computing. This picture will serve as a starting point for the design of complex neuromorphic systems targeting artificial intelligence and high-performance machine learning.
The scope of the research works presented in this Research Topic should be the reproduction of the neural phenomena involved in biological sensing (e.g., direction selectivity in the retina, echolocation) and computing (neuron integrate/fire, synaptic plasticity, pattern learning and recognition) through advanced materials, novel solid-state devices or innovative operating schemes for mainstream and emerging solid-state components. Special attention should be devoted to the physical effects and properties exploited to achieve the neuromorphic operation of the investigated artificial elements.
Topics of interest include, but are not limited to:
- advanced materials or material properties for neuromorphic hardware for sensing and computing
- novel solid-state devices for neuromorphics hardware for sensing and computing
- new physical effects in materials and solid-state devices exploitable for neuromorphic hardware for sensing and computing
- innovative operating schemes for mainstream and emerging solid-state devices allowing to reproduce principles and functionalities needed in neuromorphic hardware for sensing and computing
Papers targeting this Research Topic should provide a minimum experimental validation of the reported results (pure theoretical or simulation-only works are not suitable for the Research Topic). Research works on hardware accelerators for artificial neural networks and in-memory computing are out of the scope of the Research Topic if not closely linked to brain-inspired subjects. Research works on novel materials and devices for weight storage that do not mimic biological details are out of the scope of this Research Topic as well.
Keywords: Brain-Inspired Computing, Brain-Inspired Sensing, Physical Neural Networks, Neuromorphic Hardware, Device and Material Engineering, Device and Material Physics, Solid-State Devices
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.
