Electronic apparatus and control method thereof

Patent metadata

Field	Value
Publication number	US-11475281-B2
Application number	US-201916555331-A
Country	US
Kind code	B2
Filing date	Aug 29, 2019
Priority date	Aug 30, 2018
Publication date	Oct 18, 2022
Grant date	Oct 18, 2022

Abstract

An electronic apparatus is provided. The electronic apparatus includes a storage storing a matrix included in an artificial intelligence model, and a processor. The processor divides data included in at least a portion of the matrix by one of rows and columns of the matrix to form groups, clusters the groups into clusters based on data included in each of the groups, and quantizes data divided by the other one of rows and columns of the matrix among data included in each of the clusters.

First claim

What is claimed is:

1. An electronic apparatus comprising:
a storage storing a matrix included in an artificial intelligence model; and
a processor configured to:
divide data included in at least a portion of the matrix by one of rows and columns of the matrix to form a plurality of groups,
cluster the plurality of groups into a plurality of clusters based on data included in each of the plurality of groups, and
quantize data divided by the other one of rows and columns of the matrix among data included in each of the plurality of clusters.

2. The electronic apparatus of claim 1 ,
wherein the processor is configured to:
acquire at least one representative value based on a size of first data included in a first group and a size of second data included in a second group among groups included in each of the plurality of clusters, and convert each of the first data and the second data into a binary code based on the at least one representative value, and quantize each of the first data and the second data, and
wherein the second data is data in which the other one of rows and columns of the matrix is a same as that of the first data.

3. The electronic apparatus of claim 2 , wherein the processor is configured to:
determine a number of the at least one representative value based on a number of the plurality of groups included in each of the plurality of clusters.

4. The electronic apparatus of claim 1 , wherein the processor is configured to:
map each of the plurality of groups in a multi-dimensional space based on data included in each of the plurality of groups, and cluster the plurality of groups into the plurality of clusters based on a distance among the groups in the multi-dimensional space.

5. The electronic apparatus of claim 1 , wherein the processor is configured to:
cluster the plurality of groups into the plurality of clusters based on a K-means algorithm.

6. The electronic apparatus of claim 1 ,
wherein the processor is configured to:
based on acquiring output data for input data based on the quantized matrix, realign the output data based on clustering information, and
wherein the clustering information is acquired based on an operation of clustering the plurality of groups into the plurality of clusters.

7. The electronic apparatus of claim 1 ,
wherein the at least portion of the matrix is a sub area of the matrix determined based on the other one of rows and columns of the matrix.

8. A control method comprising:
dividing data included in at least a portion of a matrix included in an artificial intelligence model by one of rows and columns of the matrix to form a plurality of groups;
clustering the plurality of groups into a plurality of clusters based on data included in each of the plurality of groups; and
quantizing data divided by the other one of rows and columns of the matrix among data included in each of the plurality of clusters.

9. The control method of claim 8 ,
wherein the quantizing comprises:
acquiring at least one representative value based on a size of first data included in a first group and a size of second data included in a second group among groups included in each of the plurality of clusters, and converting each of the first data and the second data into a binary code based on the at least one representative value, and quantizing each of the first data and the second data, and
the second data is data wherein the other one of rows and columns of the matrix is a same as that of the first data.

10. The control method of claim 9 ,
wherein the quantizing comprises:
determining a number of the at least one representative value based on a number of the plurality of groups included in each of the plurality of clusters.

11. The control method of claim 8 ,
wherein the clustering comprises:
mapping each of the plurality of groups in a multi-dimensional space based on data included in each of the plurality of groups, and clustering the plurality of groups into the plurality of clusters based on a distance among the groups in the multi-dimensional space.

12. The control method of claim 8 ,
wherein the clustering comprises:
clustering the plurality of groups into the plurality of clusters based on a K-means algorithm.

13. The control method of claim 8 , further comprising:
based on acquiring output data for input data based on the quantized matrix, realigning the output data based on clustering information,
wherein the clustering information is acquired based on an operation of clustering the plurality of groups into the plurality of clusters.

14. The control method of claim 8 ,
wherein the at least portion of the matrix is a sub area of the matrix determined based on the other one of rows and columns of the matrix.

15. An electronic apparatus comprising:
a storage storing a matrix included in an artificial intelligence model; and
a processor configured to:
divide data in the matrix by rows to form a plurality of groups of data,
cluster the plurality of groups into a first cluster and a second cluster based on data included in the plurality of groups, and
quantize the data in the first cluster by columns according to a representative value determined for each column according to the data in the column, and quantize the data in the second cluster by columns according to a representative value determined for each column according to the data in the column.

16. The electronic apparatus of claim 15 , wherein the processor is configured to cluster the plurality of groups into the plurality of clusters by reordering at least one of the groups based on the data included in the plurality of groups.

17. The electronic apparatus of claim 15 , wherein the representative values for each column of the first cluster are different from one another, and the representative values for each column of the second cluster are different from one another.

18. The electronic apparatus of claim 15 , wherein, to quantize data in the first cluster, data from at least two columns is combined, and the combined data is quantized with one representative value.

19. The electronic apparatus of claim 15 , wherein a number of the representative values is determined based on a threshold accuracy and/or a threshold quantization error.

Assignees

• Samsung Electronics Co Ltd

Inventors

• Kapoor Parichay
• Lee Saehyung
• Lee Dongsoo
• Kim Byeoungwook

Classifications

• G06N3/06 Primary

  Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• H03M7/6041

  Compression optimized for errors · CPC title

• G06N3/08

  Learning methods · CPC title

• H03M7/3059

  Digital compression and data reduction techniques where the original information is represented by a subset or similar information, e.g. lossy compression · CPC title