US 11,983,647 B2
Method and apparatus for operating an electronic device based on a decision-making data structure using a machine learning data structure
Daniel Ansorregui Lobete, Staines (GB); and Karthikeyan Palavedu Saravanan, Staines (GB)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Appl. No. 16/468,597
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
PCT Filed Dec. 14, 2017, PCT No. PCT/KR2017/014672
§ 371(c)(1), (2) Date Jun. 11, 2019,
PCT Pub. No. WO2018/110985, PCT Pub. Date Jun. 21, 2018.
Claims priority of application No. 1621347 (GB), filed on Dec. 15, 2016.
Prior Publication US 2019/0318254 A1, Oct. 17, 2019
Int. Cl. G06N 5/045 (2023.01); G06N 20/00 (2019.01)
CPC G06N 5/045 (2013.01) [G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method for operating a first electronic device, the method comprising:
generating a decision-making data structure using a machine learning data structure, wherein the machine learning data structure includes data regarding a plurality of states, and data regarding a plurality of possible actions for each of the plurality of states, and the decision-making data structure comprises a look-up table in which each of the plurality of states is mapped with a corresponding action;
transmitting, to a second electronic device, the decision-making data structure different from the machine learning data structure;
receiving, from the second electronic device, result data regarding a result of the second electronic device performing an action that is determined based on the look-up table of the decision-making data structure; and
updating the machine learning data structure using the result data,
wherein the look-up table of the decision-making data structure includes additional information indicating data certainty of each of the plurality of states,
wherein the data certainty is represented by a first value or a second value in the look-up table, the first value indicating that more data to update the machine learning data structure is needed and the second value indicating that more data to update the machine learning data structure is not needed,
wherein the updating the machine learning data structure related to a state is performed when data certainty of the state is represented by the first value in the look-up table,
wherein the look-up table of the decision-making data structure includes a first additional column and a second additional column respectively assigned a level of data completeness and a level of data certainty,
wherein the first electronic device is configured to provide the second electronic device with four levels comprising the level of data completeness and the level of data certainty based on a combination of the first additional column and the second additional column, and
wherein, based on the data certainty of the state being represented by the first value in the look-up table, a randomly-selected action having a first specified probability level or greater is determined to be performed by the second electronic device instead of an action that is mapped to a corresponding state of the plurality of states in the look-up table.