Unlock instant, AI-driven research and patent intelligence for your innovation.

A computer-implemented neural network device and method for simulating biological neuronal networks

A neuron network and neural network technology, applied in the field of neural network organization, can solve the problems that the variables cannot represent the linear sum of independent components, and the nonlinear data processing does not satisfy the superposition principle.

Active Publication Date: 2016-05-25
ECOLE POLYTECHNIQUE FEDERALE DE LAUSANNE (EPFL)
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Nonlinear data processing does not satisfy the principle of superposition, that is, the variable to be determined cannot be expressed as a linear sum of independent components

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
  • A computer-implemented neural network device and method for simulating biological neuronal networks
  • A computer-implemented neural network device and method for simulating biological neuronal networks
  • A computer-implemented neural network device and method for simulating biological neuronal networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] FIG. 1 is a schematic diagram of a neural network device 100 . The neural network device 100 is a device that uses a system of interconnected nodes to simulate the information encoding and other processing capabilities of a biological neuronal network. The neural network device 100 can be implemented in hardware, software, or a combination of both.

[0031] The neural network device 100 includes a plurality of nodes 105 interconnected by a plurality of connections 110 . Nodes 105 are discrete information processing components similar to neurons in biological networks. Node 105 typically processes one or more input signals received via one or more connections 110 to produce one or more output signals output via one or more connections 110 . For example, in some implementations, node 105 may be an artificial neuron that weights and sums multiple input signals, substitutes the sum into one or more nonlinear activity functions, and outputs one or more output signals.

[...

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

Methods, systems and apparatus, including computer programs encoded in computer storage media, for organizing trained and untrained neural networks. In one aspect, a neural network device includes a collection of node-sets interconnected by inter-set connections, each node-set itself comprising a network of nodes interconnected by a plurality of intra-set connections, wherein each of the inter-set connections and intra-set connections is There are associated weights, each of which reflects the connection strength between the nodes connected by the associated connections, nodes in each set are more likely to be connected to other nodes in the set than to nodes in other node sets.

Description

[0001] Cross References to Related Applications [0002] Pursuant to Title 35, United States Code, Section 119(e), this application claims the benefit of U.S. Patent Application No. 61 / 301,781, filed February 5, 2010, the contents of which are incorporated herein by reference . technical field [0003] This specification relates to methods of organizing trained and untrained neural networks, and methods of organizing neural networks. Background technique [0004] Neural networks are devices inspired by structural and functional aspects of biological neuronal networks. Specifically, neural networks use a system of interconnected structures called "nodes" to mimic the information encoding and other processing capabilities of biological neuronal networks. The arrangement and strength of connections between nodes in a neural network determine the results of information processing or information storage in a neural network. [0005] Neural networks can be "trained" to produce ...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/04G06N3/08
CPCG06N3/082G06N3/045G06N3/047G06N3/063G06N3/08
Inventor H·马克拉姆R·德坎波斯佩林T·K·伯格
Owner ECOLE POLYTECHNIQUE FEDERALE DE LAUSANNE (EPFL)