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

A neural network-oriented logarithmic quantization device and method

A neural network and device technology, applied in the field of neural network-oriented logarithmic quantization devices, can solve the problems of large energy consumption, long consumption time, and prominent hardware resource consumption, so as to simplify convolution operations and improve work efficiency. Effect

Active Publication Date: 2021-07-20
INST OF COMPUTING TECH CHINESE ACAD OF SCI
View PDF13 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, with the application of neural network accelerators in terms of computing speed, resource utilization, and energy consumption, existing computing methods and data representation methods require huge multiplication and addition units, and the hardware computing process consumes a lot of energy. , and the consumption of hardware resources is prominent, and the multiplication calculation process takes a long time

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 neural network-oriented logarithmic quantization device and method
  • A neural network-oriented logarithmic quantization device and method
  • A neural network-oriented logarithmic quantization device and method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0038]The inventor found through research that the quantitative neural network data processing method can effectively avoid the multiplication operation in the convolution operation process, and the logarithmic operation of the equivalent convolution has a faster operation speed. Processing the input data of the neural network can not only greatly reduce the computing cost, hardware resources, and storage space of the convolutional neural network, but also meet the requirements of fast speed and high precision of the hardware accelerator. On the one hand, the use of very few fixed-poin...

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 provides a neural network-oriented logarithmic quantization device and a corresponding logarithmic quantization mechanism. The device realizes fast and accurate logarithmic quantization of input data by using a high-digit value extraction module and a logarithmic lookup table module, and realizes logarithmic quantization operations of input data based on a logarithmic neural network, which can be used for neural network logarithms The operation of parameterization provides logarithmic input data in preparation for further simplifying the convolution operation.

Description

technical field [0001] The present invention relates to a neural network processor architecture and a design method, specifically to the field of hardware acceleration for neural network model calculations, and more specifically to a neural network-oriented logarithmic quantization device and method. Background technique [0002] Deep learning technology has developed rapidly in recent years, and it has been widely used in solving advanced abstract cognitive problems, such as image recognition, speech recognition, natural language understanding, weather prediction, gene expression, content recommendation and intelligent robots. And has excellent performance, so it has become a research hotspot in academia and industry. Deep neural network is one of the perception models with the highest level of development in the field of artificial intelligence. This type of network simulates the neural connection structure of the human brain by building a model, and describes the data fea...

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): G06N3/063
CPCG06N3/063
Inventor 韩银和闵丰许浩博王颖
Owner INST OF COMPUTING TECH CHINESE ACAD OF SCI
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com