US 12,086,725 B2
Hyper parameter tuning for machine learning models
Suresh Kumar Golconda, Fremont, CA (US); Vijayalakshmi Krishnamurthy, Sunnyvale, CA (US); Someshwar Maroti Kale, Bangalore (IN); Sujay Sarkhel, San Jose, CA (US); Nickolas Kavantzas, Emerald Hills, CA (US); Mohan U. Kamath, Fremont, CA (US); Neelesh Kumar Shukla, Lucknow (IN); Vidya Mani, Bangalore (IN); and Amit Vaid, Bengaluru (IN)
Assigned to Oracle International Corporation, Redwood Shores, CA (US)
Filed by Oracle International Corporation, Redwood Shores, CA (US)
Filed on Aug. 6, 2020, as Appl. No. 16/987,148.
Prior Publication US 2022/0044130 A1, Feb. 10, 2022
Int. Cl. G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G06N 5/04 (2013.01) [G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
obtaining a set of universal hyper parameters for a plurality of machine learning models including a first machine learning model and a second machine learning model, wherein the set of universal hyper parameters is used to dictate a first configuration and a second configuration;
configuring a first machine learning model executing in a first computing environment in accordance with the first configuration;
configuring a second machine learning model executing independently of the first machine learning model in a second computing environment in accordance with the second configuration;
detecting, at the first computing environment, a triggering condition for tuning the set of universal hyper parameters;
based on detecting the triggering condition, adjusting a first subset of universal hyper parameters from the set of universal hyper parameters to generate a second set of universal hyper parameters;
applying the second set of universal hyper parameters to generate a third configuration; and
updating the second machine learning model in accordance with the third configuration,
wherein the method is performed by at least one device including a hardware processor.