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

A device and method for implementing rmsprop gradient descent algorithm

A gradient descent algorithm and data processing module technology, applied in machine execution devices, micro-control devices, computing, etc., can solve the problems of insufficient computing performance of general-purpose processors, reduced memory access bandwidth, and large front-end decoding overhead. IO operations, reduced bandwidth, and the effect of a high degree of parallelism

Active Publication Date: 2021-06-18
CAMBRICON TECH CO LTD
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In view of this, the main purpose of the present disclosure is to provide a device and method for executing the RMSprop gradient descent algorithm to solve the problems of insufficient computing performance of general-purpose processors for data and high decoding overhead in the previous stage, and avoid repeated reading to the memory. Fetch data and reduce the bandwidth of memory access

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 device and method for implementing rmsprop gradient descent algorithm
  • A device and method for implementing rmsprop gradient descent algorithm
  • A device and method for implementing rmsprop gradient descent algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] Other aspects, advantages and salient features of the present disclosure will become apparent to those skilled in the art from the following detailed description of the exemplary embodiments of the present disclosure in conjunction with the accompanying drawings.

[0048] In this disclosure, the terms "include" and "comprising" and their derivatives mean to include but not to limit; the term "or" is inclusive, meaning and / or.

[0049] In this specification, the various embodiments described below to describe the principles of the present disclosure are illustrative only and should not be construed as limiting the scope of the invention in any way. The following description with reference to the accompanying drawings is provided to assist in a comprehensive understanding of exemplary embodiments of the present disclosure as defined by the claims and their equivalents. The following description includes numerous specific details to aid in understanding, but these should b...

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 present disclosure provides a device and method for executing the RMSprop gradient descent algorithm. The device includes a direct memory access unit, an instruction cache unit, a controller unit, a data cache unit, and a data processing module. The method includes: first reading the gradient vector and the value vector to be updated, and at the same time initializing a mean direction vector; at each iteration, first using the gradient vector to update the mean direction vector, and then using the mean direction vector to calculate the corresponding gradient descent when updating , update the parameter vector to be updated, and repeat this process until the vector to be updated converges. During the whole process, the homogeneous vector is always stored in the data cache unit. Utilizing the disclosure, the application of the RMSprop gradient descent algorithm can be realized, and the efficiency of data processing can be greatly improved.

Description

technical field [0001] The present disclosure relates to the technical field of RMSprop algorithm application, in particular to a device and method for executing the RMSprop gradient descent algorithm, and is related to the hardware implementation of the RMSprop gradient descent optimization algorithm. Background technique [0002] Gradient descent optimization algorithm is widely used in the fields of function approximation, optimization calculation, pattern recognition and image processing. RMSprop algorithm is one of the gradient descent optimization algorithms. Features such as good effect when processing mini-batch data sets are widely used, and using a dedicated device to implement the RMSprop algorithm can significantly increase its execution speed. [0003] Currently, one known way to implement the RMSprop gradient descent algorithm is to use a general-purpose processor. The method supports the above-mentioned algorithms by executing general-purpose instructions usi...

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): G06F9/30G06F9/22
CPCG06F9/223G06F9/30003G06F9/30145
Inventor 刘少礼郭崎陈天石陈云霁
Owner CAMBRICON TECH CO LTD
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com