US 11,985,527 B2
Systems and methods for autonomous network management using deep reinforcement learning
Said Soulhi, Boston, MA (US); and Bryan Christopher Larish, Westfield, NJ (US)
Assigned to Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed by Verizon Patent and Licensing Inc., Basking Ridge, NJ (US)
Filed on Jan. 20, 2023, as Appl. No. 18/157,469.
Application 18/157,469 is a continuation of application No. 17/101,749, filed on Nov. 23, 2020, granted, now 11,601,830.
Prior Publication US 2023/0156501 A1, May 18, 2023
Int. Cl. H04W 24/08 (2009.01); G06N 3/045 (2023.01); G06N 3/088 (2023.01); G06N 10/60 (2022.01); H04L 41/16 (2022.01); H04L 43/0817 (2022.01); H04W 24/02 (2009.01); H04W 28/02 (2009.01)
CPC H04W 24/08 (2013.01) [G06N 3/045 (2023.01); G06N 3/088 (2013.01); G06N 10/60 (2022.01); H04L 41/16 (2013.01); H04L 43/0817 (2013.01); H04W 24/02 (2013.01); H04W 28/0268 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A device, comprising:
one or more processors configured to:
generate a plurality of sets of encoded parameters by encoding a set of parameters in a plurality of different dimensions, wherein the plurality of different dimensions include at least a first dimension and a second dimension, wherein the plurality of sets of encoded parameters include:
a first set of encoded parameters encoded in a first dimension, wherein the first set of encoded parameters includes a first quantity of values to represent the set of parameters, and
a second set of encoded parameters encoded in a second dimension, wherein the second dimension includes a second quantity of values to represent the set of parameters;
decode the plurality of sets of encoded parameters to generate a plurality of sets of decoded parameters;
calculate a plurality of reconstruction errors associated with the plurality of sets of decoded parameters and the set of parameters, wherein each respective reconstruction error of the plurality of reconstruction errors is calculated based on differences between a respective set of decoded parameters and the set of parameters;
calculate a multi-dimension reconstruction error based on the calculated plurality of reconstruction errors; and
modify a configuration of a wireless network based on the calculated multi-dimension reconstruction error.