US 12,223,411 B2
Electronic apparatus and method of controlling thereof
Chiyoun Park, Suwon-si (KR); Jaedeok Kim, Suwon-si (KR); Hyunjoo Jung, Suwon-si (KR); and Inkwon Choi, Suwon-si (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Appl. No. 17/053,005
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
PCT Filed May 30, 2019, PCT No. PCT/KR2019/006500
§ 371(c)(1), (2) Date Nov. 4, 2020,
PCT Pub. No. WO2019/245186, PCT Pub. Date Dec. 26, 2019.
Claims priority of application No. 10-2018-0070374 (KR), filed on Jun. 19, 2018.
Prior Publication US 2021/0125028 A1, Apr. 29, 2021
Int. Cl. G06N 3/04 (2023.01); G06N 3/08 (2023.01)
CPC G06N 3/04 (2013.01) [G06N 3/08 (2013.01)] 8 Claims
OG exemplary drawing
 
1. A method of controlling an electronic apparatus for generating an artificial intelligence model, the method comprising:
based on receiving, by the electronic apparatus, a user command for using the artificial intelligence model, identifying, by the electronic apparatus, a memory capacity of the electronic apparatus;
based on the identified memory capacity of the electronic apparatus, identifying, by the electronic apparatus, a number of artificial intelligence models;
generating, by the electronic apparatus, a plurality of artificial intelligence models based on the identified number of the artificial intelligence models;
obtaining, by the electronic apparatus, output data by inputting input data to the plurality of artificial intelligence models; and
outputting, by the electronic apparatus, a result value by ensembling the output data obtained from the plurality of artificial intelligence models,
wherein the generating of the plurality of artificial intelligence models includes:
receiving, from a server, a reference value for obtaining a plurality of layer parameters with respect to layers at corresponding locations among a plurality of layers included in each of the plurality of artificial intelligence models without receiving, from the server, the plurality of artificial intelligence models, and
based on the received reference value, generating the layers at corresponding locations among the plurality of layers included in each of the plurality of artificial intelligence models,
wherein the generating of the plurality of artificial intelligence models further includes generating a layer scheme of the plurality of artificial intelligence models based on the received reference value,
wherein the reference value is determined based on an average value and a distribution value of the plurality of layer parameters with respect to the layers at corresponding locations, and
wherein, based on the distribution value of the plurality of layer parameters being less than or equal to a predetermined value, the reference value is determined based on the average value of the plurality of layer parameters.