Machine learning system and method and apparatus for creating machine learning system

A machine learning and filter coefficient technology, applied in machine learning, neural learning methods, instruments, etc., can solve the problem that RNN cannot be strongly quantized, and achieve the effect of saving energy and reducing filter coefficients

Pending Publication Date: 2020-11-10
ROBERT BOSCH GMBH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the downside of RNNs is that these RNNs cannot be hardened

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
  • Machine learning system and method and apparatus for creating machine learning system
  • Machine learning system and method and apparatus for creating machine learning system
  • Machine learning system and method and apparatus for creating machine learning system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] figure 1 A schematic diagram of a diagram is shown in which the measured energy consumption E of a neural network (NN-temp) with temporal filtering and the measured energy consumption E of a RNN are plotted against the quantization resolution Q. Note that both networks achieve similar results on the same dataset. The y-axis illustrates energy consumption E, which is lowest at the origin of the graph and increases with distance from the origin. The quantization resolution Q is plotted on the x-axis, increasing towards the origin of the graph. That is, the closer to the origin, the higher the quantization resolution, especially the number of bits used to represent the parameters. In the following, an exemplary case is considered: a neural network with temporal filtering (NN-temp) has a significantly greater energy consumption for high quantization resolutions than a similar RNN at the same quantization resolution.

[0051] If the resolution of the parameters of the two...

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

A machine learning system and a method and apparatus for creating a machine learning system are provided. The invention relates to the machine learning system (10) comprising at least one time filter(21). An input variable comprising a temporal sequence of images is processed using the filter (21) by means of a machine learning system (10). The machine learning system (10) is configured to applythe filter (21) to a sequence of pixels each at a respective same coordinate of the image or at a respective same coordinate of an intermediate result. The filter coefficients of the filter (21) are quantized. Furthermore, the invention relates to a method, a computer program and a device for creating a machine learning system (10).

Description

technical field [0001] The present invention relates to a machine learning system with quantized parameters. Furthermore, the invention relates to methods and computer programs and devices for creating machine learning systems. Background technique [0002] The non-prepublished document DE 10 2017 218 889.8 discloses an artificial intelligence module which is designed to process one or more input variables into one or more output variables via an internal processing chain. The internal processing chain is specified by one or more parameters. An assignment module is provided, which is designed to determine the parameter as a function of at least one statistical distribution. [0003] The non-prepublished document DE 10 2018 216 471.1 discloses a quantized neural network. [0004] It is possible to create quantized neural networks with extremely strong quantization, as e.g. by Hubara et al. "Quantized neural networks: Training neural networks with low precisionweights and a...

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 Applications(China)
IPC IPC(8): G06N3/04G06N3/063G06N3/08
CPCG06N3/04G06N3/08G06N3/063G06N3/048G06N3/045G06N20/00G05D1/0088G06T7/00
Inventor T.普法伊尔
Owner ROBERT BOSCH GMBH
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