US 12,450,172 B2
Reducing set of policies with two or more hyper-dimensions
Amir Rosen, Haifa (IL); Ariel Szapiro, Kfar Neter (IL); Gil Levy, Hod Hasharon (IL); Arye Albahari, Kiryat Motzkin (IL); and Sagi Lahav, Kiryat Bialik (IL)
Assigned to Mellanox Technologies, Ltd, Yokneam (IL)
Filed by Mellanox Technologies, Ltd., Yokneam (IL)
Filed on Apr. 1, 2024, as Appl. No. 18/623,103.
Prior Publication US 2025/0307177 A1, Oct. 2, 2025
Int. Cl. G06F 12/12 (2016.01); G06F 12/08 (2016.01); G06F 12/0862 (2016.01); G06F 12/126 (2016.01)
CPC G06F 12/126 (2013.01) [G06F 12/0862 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
finding reduced sets of policies for respective hyper-dimensions including finding a first reduced set of policies for a first hyper-dimension and finding a second reduced set of policies for a second hyper-dimension;
combining the reduced sets of policies in different combinations yielding a third set of policies including combining the first reduced set of policies for the first hyper-dimension and the second reduced set of policies of policies for the second hyper-dimension; and
reducing the third set of policies to yield a fourth reduced set of policies available for selection by a machine learning agent, each policy in the fourth reduced set including an array of values associated with each of the hyper-dimensions including an array of first values associated with the first hyper-dimension, and an array of second values associated with the second hyper-dimension.