With the improvement in data collection and computing power, artificial intelligence (AI), represented by deep learning, has been developing rapidly. However, with the increasing model complexity, the model becomes more and more dependent on amounts of high-quality data, while its interpretability remains standstill. In many applications, especially in computer-aided diagnosis (CAD), the standard data are mainly unbalanced and small-size. Therefore, traditional learning methods based on large samples cannot be directly and effectively applied in CAD. It usually manifests in weak generalization ability, poor interpretability, and weak robustness to noise.
On the other hand, humans are good at learning from small samples, and can draw inferences by analogy. Therefore, inspired by the human learning ability, how to quickly learn based on small samples to form brain-like intelligence has become the core issue of the development of a new generation of AI.
Whether it is the improvement of clinical diagnostic ability or the development of the basic theoretical system of the new generation of AI, it is urgent to conduct in-depth research on small sample learning, especially, the learning mechanism and mathematical mechanism of small samples. It will not only help to improve the ability of early diagnosis and intervention of clinical diseases, but also help to establish an understandable and explicable learning mechanism. Therefore, in this research topic, we aim to establish a bridge between brain functional analysis and brain-like machine intelligence, which will promote the basic theory of AI, as well as the mechanism of brain function.
This research topic aims to provide a bridge between the brain's functional analysis and brain-like machine intelligence. Specifically, we solicit original contributions related to the following three areas: (1) Brain structure and functional analysis; (2) Brain simulation; and (3) Brain-inspired learning methods with applications. Submissions are encouraged to cover any of the above-mentioned questions from every possible perspective. Topics of interest include (but are not limited to):
· Multimodal data-based brain structural and functional analysis
· Simulation of neuron / neural dynamics / visual system/mechanism of human learning
· Small sample learning/transfer learning/reinforcement learning
· Computational methodology for AI
· Applications in computer-aided diagnosis/computer vision
With the improvement in data collection and computing power, artificial intelligence (AI), represented by deep learning, has been developing rapidly. However, with the increasing model complexity, the model becomes more and more dependent on amounts of high-quality data, while its interpretability remains standstill. In many applications, especially in computer-aided diagnosis (CAD), the standard data are mainly unbalanced and small-size. Therefore, traditional learning methods based on large samples cannot be directly and effectively applied in CAD. It usually manifests in weak generalization ability, poor interpretability, and weak robustness to noise.
On the other hand, humans are good at learning from small samples, and can draw inferences by analogy. Therefore, inspired by the human learning ability, how to quickly learn based on small samples to form brain-like intelligence has become the core issue of the development of a new generation of AI.
Whether it is the improvement of clinical diagnostic ability or the development of the basic theoretical system of the new generation of AI, it is urgent to conduct in-depth research on small sample learning, especially, the learning mechanism and mathematical mechanism of small samples. It will not only help to improve the ability of early diagnosis and intervention of clinical diseases, but also help to establish an understandable and explicable learning mechanism. Therefore, in this research topic, we aim to establish a bridge between brain functional analysis and brain-like machine intelligence, which will promote the basic theory of AI, as well as the mechanism of brain function.
This research topic aims to provide a bridge between the brain's functional analysis and brain-like machine intelligence. Specifically, we solicit original contributions related to the following three areas: (1) Brain structure and functional analysis; (2) Brain simulation; and (3) Brain-inspired learning methods with applications. Submissions are encouraged to cover any of the above-mentioned questions from every possible perspective. Topics of interest include (but are not limited to):
· Multimodal data-based brain structural and functional analysis
· Simulation of neuron / neural dynamics / visual system/mechanism of human learning
· Small sample learning/transfer learning/reinforcement learning
· Computational methodology for AI
· Applications in computer-aided diagnosis/computer vision