Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Neural network weight discretization method and system

A neural network and discretization technology, applied in the field of neural network weight discretization, to achieve the effect of reducing computational complexity and storage space consumption

Active Publication Date: 2019-01-15
TSINGHUA UNIV
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Based on this, it is necessary to provide a neural network weight discretization method and system for how to balance the performance and computational complexity of the neural network. The method includes:

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
  • Neural network weight discretization method and system
  • Neural network weight discretization method and system
  • Neural network weight discretization method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0069] 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 in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0070] The main operation part of the neural network calculation is the multiplication and accumulation operation of the input data and the weight, and the accuracy of the weight directly affects the complexity of the calculation. The present invention can effectively reduce the accuracy requirement of the weight through the method of probability sampling, significantly reduce the memory usage of the weight memory and reduce the operation complexity under the premise of almost no influence on the performance. This method can be used not only for offline learning of neural networks, but also for on-...

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 relates to a neural network weight discretization method and system. The method comprises the following steps: obtaining a weight value range and a number of discrete weight states; acquiring a previous time step weight state and a current time step weight increment; obtaining a state transition direction by using a direction function according to the weight increment of the currenttime step; obtaining a current time step weight state according to the previous time step weight state, the current time step weight increment, the state transition direction, the weight value range,and the number of discrete weight states. The invention ensures that the weight value is always constrained in the same discrete state space without storing additional virtual continuous state implicit weights, and greatly reduces the consumption of storage space and the computational complexity under the premise of ensuring the computational performance of the neural network.

Description

technical field [0001] The invention relates to the technical field of neural networks, in particular to a method and system for discretizing neural network weights. Background technique [0002] In the past few years, large data sets, different learning models, and GPUs (General Purpose Graphic Processing Units) have made deep neural networks more popular in artificial intelligence fields such as computer vision, speech recognition, natural language processing, and human-computer Go games. One result after another has been achieved. However, behind these striking results, there is also a huge overhead of hardware resources, training time overhead and power consumption required to calculate them. Therefore, GPU-based deep learning systems are difficult to embed in portable devices. However, many solutions with high power consumption utility often require a large loss in performance. Therefore, there is a trade-off between performance and computational complexity. [0003...

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): G06N3/04
CPCG06N3/04G06N3/084G06N3/063G06N3/045G06N3/08G06N7/01
Inventor 李国齐邓磊吴臻志裴京
Owner TSINGHUA UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products