Method and system for constructing adaptive neural network model based on brain development mechanism
A neural network model and construction method technology, applied in biological neural network models, neural learning methods, neural architectures, etc., can solve the problems of time-consuming computing, large search space, high computing resources and hardware requirements, and achieve dynamic allocation, The effect of improving accuracy
Inactive Publication Date: 2020-02-07
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF12 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
This method usually has a huge search space, is time-consuming to calculate, and has very high requirements on computing resources and hardware.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0068] Preferred embodiments of the present invention are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention, and are not intended to limit the protection scope of the present invention.
[0069] The purpose of the present invention provides a method for building an adaptive neural network model based on the brain development mechanism. According to the number of neurons and connection weights of the artificial neural network, the neurons are pruned, and a certain proportion of unimportant ones are cut off. After adding neurons, train the network again until the adaptability of the network reaches the highest value, so that the dynamic allocation of neurons can be realized, and the accuracy of sample classification can be improved.
[0070] Among them, learning from the working principle of the brain nervous system is an effecti...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention relates to a method and system for constructing an adaptive neural network model based on a brain development mechanism.. The method comprises the steps of initializing a fully-connectedthree-layer artificial neural network; according to the number of neurons of the artificial neural network and the connection weight, pruning the neurons to obtain a pruned network; retraining the pruning network by adopting a back propagation algorithm to obtain an updated network; calculating an adaptability value of the updated network; and adjusting the pruning condition of neurons accordingto the adaptability value so as to obtain an updated network corresponding to the optimal adaptability value, the updated network corresponding to the optimal adaptability value being the adaptive neural network model. According to the number and the connection weight of the neurons of the artificial neural network, after the neurons are pruned and a certain proportion of unimportant neurons are cut off, the network is trained again until the adaptability of the network reaches the highest value, so that the dynamic distribution of the neurons can be realized, and the accuracy of sample classification is further improved.
Description
technical field [0001] The invention relates to the technical field of computational neuroscience, in particular to a method and system for constructing an adaptive neural network model based on a brain development mechanism. Background technique [0002] In recent years, discoveries from neuroscience have influenced artificial intelligence research from several perspectives. For example, deep neural networks draw on the hierarchical information processing mechanism of the brain's visual system; long-term and short-term memory networks draw on the memory and forgetting of the human brain in information processing. Mechanism, through the gating mechanism, the input information is retained or deleted to better retain key information. However, the training methods of these networks are non-biological methods, and the learning process often requires a large number of manual annotations. [0003] In addition, the current neural network needs to define the network structure in ad...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06N3/04G06N3/08
CPCG06N3/084G06N3/082G06N3/045
Inventor 赵菲菲张铁林曾毅
Owner INST OF AUTOMATION CHINESE ACAD OF SCI