Brain-inspired intelligence is a kind of machine intelligence, which is inspired by brain neural mechanism and cognitive behavior mechanism by means of computational modeling and realized by the cooperation of software and hardware. Digital neuromorphic system is based on spiking neural network (SNN) models or hybrid SNN models. Its goal is to make the machine realize all kinds of human cognitive ability based on brain-inspired mechanisms with online learning performance, thus reaching or surpassing the level of human intelligence.
This Research Topic focuses on digital neuromorphic systems with online learning performance, including algorithms and its implementation. There has been some recent progress of digital neuromorphic systems towards artificial general intelligence based on the following facts. A series of spiking neural network algorithms are developed for various types of applications. It significantly facilitates the development of the neuromorphic computing, especially digital neuromorphic systems. A few learning algorithms are presented, aiming at lifelong learning or online learning of digital neuromorphic systems. Large-scale architecture design can further improve the capability of digital neuromorphic systems. In addition, analog circuits can be integrated into the digital neuromorphic systems to realize a mixed brain-inspired system. Therefore, the gap now is how to integrate these facts to improve the performance of digital neuromorphic systems.
Relevant sub-topics of this Research Topic include but are not limited to the following:
- Novel SNNs with online learning capability for digital neuromorphic computing
- Online learning architecture for large-scale digital neuromorphic systems
- Online learning algorithms for lifelong adaptation of digital neuromorphic systems, including supervised learning, meta-learning, unsupervised learning, etc.
- Deep SNN models and hybrid SNN models with novel online learning algorithms
- Novel digital neuromorphic chips or systems, and digital neuromorphic component for mixed analog-digital neuromorphic computing
- Spike-based perception algorithms or systems based on digital neuromorphic computing, including vision and auditory perception
- Adaptive intelligent robot platform using digital neuromorphic systems
Brain-inspired intelligence is a kind of machine intelligence, which is inspired by brain neural mechanism and cognitive behavior mechanism by means of computational modeling and realized by the cooperation of software and hardware. Digital neuromorphic system is based on spiking neural network (SNN) models or hybrid SNN models. Its goal is to make the machine realize all kinds of human cognitive ability based on brain-inspired mechanisms with online learning performance, thus reaching or surpassing the level of human intelligence.
This Research Topic focuses on digital neuromorphic systems with online learning performance, including algorithms and its implementation. There has been some recent progress of digital neuromorphic systems towards artificial general intelligence based on the following facts. A series of spiking neural network algorithms are developed for various types of applications. It significantly facilitates the development of the neuromorphic computing, especially digital neuromorphic systems. A few learning algorithms are presented, aiming at lifelong learning or online learning of digital neuromorphic systems. Large-scale architecture design can further improve the capability of digital neuromorphic systems. In addition, analog circuits can be integrated into the digital neuromorphic systems to realize a mixed brain-inspired system. Therefore, the gap now is how to integrate these facts to improve the performance of digital neuromorphic systems.
Relevant sub-topics of this Research Topic include but are not limited to the following:
- Novel SNNs with online learning capability for digital neuromorphic computing
- Online learning architecture for large-scale digital neuromorphic systems
- Online learning algorithms for lifelong adaptation of digital neuromorphic systems, including supervised learning, meta-learning, unsupervised learning, etc.
- Deep SNN models and hybrid SNN models with novel online learning algorithms
- Novel digital neuromorphic chips or systems, and digital neuromorphic component for mixed analog-digital neuromorphic computing
- Spike-based perception algorithms or systems based on digital neuromorphic computing, including vision and auditory perception
- Adaptive intelligent robot platform using digital neuromorphic systems