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

Method for executing multiplication operation by neural network processor and neural network processor

A neural network and multiplication technology, applied in biological neural network models, electrical digital data processing, digital data processing components, etc., can solve problems such as consumption, improper use of processing resources, large chip area and power

Pending Publication Date: 2022-05-27
SAMSUNG ELECTRONICS CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Processors used for neural networks can perform a large number of multiplications and additions, some of which may be important to Dealing with resource misuse
Furthermore, some operations in such systems may be integer operations and some operations may be floating point operations, which may consume significant die area and power if performed on separate sets of dedicated hardware

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
  • Method for executing multiplication operation by neural network processor and neural network processor
  • Method for executing multiplication operation by neural network processor and neural network processor
  • Method for executing multiplication operation by neural network processor and neural network processor

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The detailed description set forth below in connection with the accompanying drawings is intended as a description of exemplary embodiments of processors for fine-grain sparse integer and floating point operations provided in accordance with the present disclosure, and is not intended to represent the present The only form in which the disclosure can be constructed or utilized. The description sets forth the features of the disclosure in conjunction with the illustrated embodiments. It should be understood, however, that the same or equivalent functions and structures may be implemented by different embodiments, which are also intended to be included within the scope of the disclosure. As indicated elsewhere herein, the same element numbers are intended to refer to the same elements or features.

[0032] Neural networks (eg, when performing inference) can perform a large number of computations in which activations (or "activation values") (elements of an input feature ...

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 discloses a method for executing multiplication by a neural network processor and the neural network processor. In some embodiments, the method includes forming a first set of products and forming a second set of products. The step of forming the first set of products may include: multiplying a first activation value with a least significant sub-word and a most significant sub-word of a first weight in a first multiplier to form a first partial product and a second partial product; and adding the first partial product to the second partial product. The step of forming the second set of products may include: multiplying the second activation value with the first and second sub-words of the mantissa in the first multiplier to form a third partial product and a fourth partial product; and adding the third partial product to the fourth partial product.

Description

[0001] This application claims a document entitled "SYSTEM AND METHOD FOR IMPROVING AREA AND POWEREFFICIENCY BY REDISTRIBUTING WEIGHT NIBBLES", filed November 11, 2020, entitled "SYSTEM AND METHOD FOR IMPROVING AREA AND POWEREFFICIENCY BY REDISTRIBUTING WEIGHT NIBBLES" Priority to and benefit from US Provisional Application No. 63 / 112,271, and US Application No. 17 / 131,357, filed December 22, 2020, the entire contents of which are incorporated herein by reference. technical field [0002] One or more aspects of embodiments in accordance with the present disclosure relate to processing circuits, and more particularly, to systems and methods for performing sets of multiplications in a manner that accommodates outliers and is capable of performing both integer and floating point operations. Background technique [0003] A processor used in a neural network can perform a large number of multiplication and addition operations, and some of the multiplication and addition operations...

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): G06N3/063G06F7/487G06F7/544
CPCG06N3/063G06F7/4876G06F7/5443G06F5/012G06F7/523G06F7/485G06F7/4836
Inventor 阿里·沙菲·阿得斯塔尼约瑟夫·哈松
Owner SAMSUNG ELECTRONICS CO LTD