US 12,001,932 B2
Hierarchical reinforcement learning algorithm for NFV server power management
Zhu Zhou, Portland, OR (US); Xiaotian Gao, Beijing (CN); Chris MacNamara, Limerick (IE); Stephen Doyle, Ennis (IE); and Atul Kwatra, Gilbert, AZ (US)
Assigned to Intel Corporation, Santa Clara, CA (US)
Filed by Intel Corporation, Santa Clara, CA (US)
Filed on Jul. 27, 2020, as Appl. No. 16/939,237.
Prior Publication US 2020/0356834 A1, Nov. 12, 2020
Int. Cl. G06N 3/006 (2023.01); G06F 1/3287 (2019.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G06N 3/006 (2013.01) [G06F 1/3287 (2013.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01)] 23 Claims
OG exemplary drawing
 
1. A method implemented in one or more systems, each including a processor having a core comprising a first portion of circuitry including a plurality of cores and a second portion of circuitry external to the core, the method comprising:
while executing software on a first processor in a first system to perform a first workload,
training a first machine learning (ML) model by adjusting a frequency of the core of the first processor to obtain a first trained ML model; and
implementing the trained ML model in an inference mode while training a second ML model by adjusting a frequency of the core and a frequency of the second portion of circuitry of the first processor to obtain a second trained ML model.