US 12,373,428 B2
Machine learning models in an artificial intelligence infrastructure
Brian Gold, Los Altos, CA (US); Emily Watkins, Houston, TX (US); Ivan Jibaja, San Jose, CA (US); Igor Ostrovsky, Mountain View, CA (US); and Roy Kim, Los Altos, CA (US)
Assigned to PURE STORAGE, INC., Santa Clara, CA (US)
Filed by PURE STORAGE, INC., Mountain View, CA (US)
Filed on Apr. 26, 2022, as Appl. No. 17/660,735.
Application 17/660,735 is a continuation of application No. 16/515,698, filed on Jul. 18, 2019, granted, now 11,403,290.
Application 16/515,698 is a continuation of application No. 16/045,814, filed on Jul. 26, 2018, granted, now 10,360,214, issued on Jul. 23, 2019.
Claims priority of provisional application 62/650,736, filed on Mar. 30, 2018.
Claims priority of provisional application 62/648,368, filed on Mar. 26, 2018.
Claims priority of provisional application 62/620,286, filed on Jan. 22, 2018.
Claims priority of provisional application 62/579,057, filed on Oct. 30, 2017.
Claims priority of provisional application 62/576,523, filed on Oct. 24, 2017.
Claims priority of provisional application 62/574,534, filed on Oct. 19, 2017.
Prior Publication US 2022/0253443 A1, Aug. 11, 2022
Int. Cl. G06F 3/06 (2006.01); G06F 16/22 (2019.01); G06F 16/2453 (2019.01); G06F 18/213 (2023.01); G06N 3/08 (2023.01); G06N 20/00 (2019.01); G06T 1/20 (2006.01); G06T 1/60 (2006.01)
CPC G06F 16/24534 (2019.01) [G06F 3/06 (2013.01); G06F 3/061 (2013.01); G06F 3/0629 (2013.01); G06F 3/0647 (2013.01); G06F 16/2255 (2019.01); G06F 18/213 (2023.01); G06N 3/08 (2013.01); G06N 20/00 (2019.01); G06T 1/20 (2013.01); G06T 1/60 (2013.01); G06T 2200/28 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
storing, within one or more storage systems of an artificial intelligence infrastructure, information describing a dataset and one or more transformations applied to the dataset resulting in a transformed dataset;
identifying, by the artificial intelligence infrastructure, previous versions of a machine learning model that used the transformed dataset as input during one or more prior executions by the artificial intelligence infrastructure;
storing, within the one or more storage systems, information describing only differences between the previous versions of the machine learning model and a current version of the machine learning model;
retaining one or more first portions of the transformed dataset associated with the current version of the machine learning model at a first storage tier; and
moving one or more second portions of the transformed dataset associated with the previous versions of the machine learning model to a second storage tier.