Beyond the great success in machine learning (ML) based artificial intelligence, the engineering community has long been learning and imitating biological systems from almost every aspect (e.g., neural networks, robots) to seek artificial general intelligence (AGI). The human brain and sensory system are a sophisticated biological machine based on hierarchical nervous meshes, interconnecting “slow” and “inaccurate” neural components (notably neurons and synapses). Yet, it is capable to precisely perceive the environment, instantly calculate and react accordingly in real-time and low power, which is far more efficient and faster than any software/hardware computing platform can accomplish for the given level of scale. Therefore, biologically plausible implementations have been an important inspiration for acquiring external information (e.g., sensors), and more importantly, processing and executing the merged external/internal information through a cognitive process (e.g., neuromorphic computing systems), which is a long-standing goal, i.e. AGI for the engineering communities to achieve. In terms of hardware implementation, the researches on brain-inspired computing including new concept devices, circuits and chips have attracted a great interest.
The state-of-the-art advances of the brain-inspired artificial innb
Benefiting from the constant interactive dialogue between engineering and neuroscience, innovative solutions have been put forward from each hierarchical level, from device, circuit, architecture to algorithm. A notable example concerning algorithm is the spiking neural network (SNN) that shows superior advantage in terms of speed and power efficiency when processing complex spatio-temporal data with additional noise and sparsity in the event-driven computing paradigm. On the device side, emerging devices with multiple internal dynamics (memristive, phase-change, spintronic, etc.) are being researched to implement artificial synapses or neurons. Accordingly, circuitry and architecture interfacing and orchestrating these systems are indispensable and will further leverage the overall performance. Consequently, synergy from the advent of bio-plausible device technology, well-established practices of circuit/architecture-oriented approach, to the biological algorithm is essential and will cast a significant impact on the progress of the practical AGI.
This Research Topic aims to bring together the state-of-the-art interdisciplinary research from a point views of hardware implementation in the above context involving enabling technologies for brain-inspired from the perspectives of both the engineering and neuroscience field including, but not limited to biological principles and engineering approaches for intelligent sensing and computing, with emphasis on brain-inspired emerging devices, circuit, and architecture. Welcome contributions include:
• Emerging technologies and devices emulating multiple neural functions (e.g., synapses, neurons) for brain-inspired processing systems (sensory and computing components and systems)
• Emerging circuitry and architectural designs for exploiting the neuromorphic devices and exploring the brain-inspired hierarchical systems
Beyond the great success in machine learning (ML) based artificial intelligence, the engineering community has long been learning and imitating biological systems from almost every aspect (e.g., neural networks, robots) to seek artificial general intelligence (AGI). The human brain and sensory system are a sophisticated biological machine based on hierarchical nervous meshes, interconnecting “slow” and “inaccurate” neural components (notably neurons and synapses). Yet, it is capable to precisely perceive the environment, instantly calculate and react accordingly in real-time and low power, which is far more efficient and faster than any software/hardware computing platform can accomplish for the given level of scale. Therefore, biologically plausible implementations have been an important inspiration for acquiring external information (e.g., sensors), and more importantly, processing and executing the merged external/internal information through a cognitive process (e.g., neuromorphic computing systems), which is a long-standing goal, i.e. AGI for the engineering communities to achieve. In terms of hardware implementation, the researches on brain-inspired computing including new concept devices, circuits and chips have attracted a great interest.
The state-of-the-art advances of the brain-inspired artificial innb
Benefiting from the constant interactive dialogue between engineering and neuroscience, innovative solutions have been put forward from each hierarchical level, from device, circuit, architecture to algorithm. A notable example concerning algorithm is the spiking neural network (SNN) that shows superior advantage in terms of speed and power efficiency when processing complex spatio-temporal data with additional noise and sparsity in the event-driven computing paradigm. On the device side, emerging devices with multiple internal dynamics (memristive, phase-change, spintronic, etc.) are being researched to implement artificial synapses or neurons. Accordingly, circuitry and architecture interfacing and orchestrating these systems are indispensable and will further leverage the overall performance. Consequently, synergy from the advent of bio-plausible device technology, well-established practices of circuit/architecture-oriented approach, to the biological algorithm is essential and will cast a significant impact on the progress of the practical AGI.
This Research Topic aims to bring together the state-of-the-art interdisciplinary research from a point views of hardware implementation in the above context involving enabling technologies for brain-inspired from the perspectives of both the engineering and neuroscience field including, but not limited to biological principles and engineering approaches for intelligent sensing and computing, with emphasis on brain-inspired emerging devices, circuit, and architecture. Welcome contributions include:
• Emerging technologies and devices emulating multiple neural functions (e.g., synapses, neurons) for brain-inspired processing systems (sensory and computing components and systems)
• Emerging circuitry and architectural designs for exploiting the neuromorphic devices and exploring the brain-inspired hierarchical systems