Feedforward sequence memory neuron network and construction method and system thereof

A neural network and memory technology, applied in the field of artificial intelligence, can solve problems such as the inability of neural networks to guarantee information processing efficiency.

Active Publication Date: 2017-07-04
IFLYTEK CO LTD
View PDF2 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The embodiment of the present invention provides a feedforward sequence memory neural network and its construction method and system, which solves the problem that the existing neural n

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Feedforward sequence memory neuron network and construction method and system thereof
  • Feedforward sequence memory neuron network and construction method and system thereof
  • Feedforward sequence memory neuron network and construction method and system thereof

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0071] Example one

[0072] A feedforward sequence memory neural network includes multiple nodes of at least three layers, the first layer is the input layer, the last layer is the output layer, and the other multiple nodes located between the input layer and the output layer form at least one hidden layer, The nodes between the layers are fully connected, and also include: each hidden layer contains a memory block, the hidden layer and the memory block together form a bidirectional feedforward sequential memory neural network FSMN layer, where the memory block of the current hidden layer The input of is the output of the current hidden layer, the output of the current hidden layer memory block is an input of the next hidden layer, the memory block is used to store the historical information and future information of each frame of input data, and the historical information is the current frame The characteristic sequence of the frame before the input data, the future information ...

Example Embodiment

[0074] Example two

[0075] A feedforward sequential memory neural network, as described in the first embodiment, the difference is that in this embodiment, in order to improve the ability of the neural network to process information and data, the bidirectional FSMN stack is replaced with a bidirectional LSFSMN stack, Each bidirectional LSFSMN layer is composed of a bidirectional FSMN layer and an LSTM layer in the same layer, wherein the LSTM layer is used to memorize historical information, and the bidirectional FSMN layer is used to memorize future information. The neural network structure has better memory capabilities for input data than the neural network structure provided in the first embodiment.

[0076] A feedforward sequence memory neural network includes multiple nodes of at least three layers, the first layer is the input layer, the last layer is the output layer, and the other multiple nodes between the input layer and the output layer form at least one hidden layer, ...

Example Embodiment

[0077] Example three

[0078] A feedforward sequence memory neural network, as described in the first embodiment, the difference is that in this embodiment, in order to improve the ability of the neural network to process information and data, the neural network structure further includes a fully connected stack. The information processing capability of the neural network structure is better than the neural network structure provided in the first embodiment, and the information processing efficiency will not decrease significantly.

[0079] A feedforward sequence memory neural network includes multiple nodes of at least three layers, the first layer is the input layer, the last layer is the output layer, and the other multiple nodes between the input layer and the output layer form at least one hidden layer, The nodes between the layers are fully connected. Each hidden layer contains a memory block. The hidden layer and the memory block together form the bidirectional feedforward s...

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

No PUM Login to view more

Abstract

The invention discloses a feedforward sequence memory neuron network and a construction method and system thereof. The feedforward sequence memory neuron network comprises at least three layers of multiple nodes, wherein the first layer is an input layer, the last layer is an output layer, and other multiple nodes positioned between the input layer and the output layer form at least one hidden layer; each hidden layer comprises a memory block; the hidden layer and the memory block jointly form a bidirectional FSMN (Feedforward Sequence Memory Neuron Network) layer; and the memory block is used for storing the historical information and the future information of each frame of input information. Since the bidirectional FSMN layer comprises the memory block, the historical information and the future information of each frame of input information are stored, the long-time information of training data can be utilized, the process does not need to carry out bidirectional circulation feedforward, and information processing efficiency can be guaranteed.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a feedforward sequential memory neural network and its construction method and system. Background technique [0002] Artificial neural network is a kind of model established by simulating the human brain nervous system from the perspective of microstructure and function. It has the ability to simulate part of human image thinking, and its characteristics are mainly nonlinear characteristics, learning ability and adaptability. , is an important way to realize artificial intelligence. It is a network composed of simple information processing units interconnected, which can accept and process information. The information processing of the network is realized by the interaction between units. Handled for connection weights between processing units. In recent years, neural networks have played a vital role in the application systems of human-computer interaction, such as speech...

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
no application Login to view more
IPC IPC(8): G06N3/04
CPCG06N3/044
Inventor 张仕良熊世富魏思潘嘉刘聪胡国平胡郁刘庆峰
Owner IFLYTEK CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap