Intermediate network node and method performed therein for handling data of communication networks

Patent metadata

Field	Value
Publication number	US-2022172054-A1
Application number	US-201917598474-A
Country	US
Kind code	A1
Filing date	Mar 28, 2019
Priority date	Mar 28, 2019
Publication date	Jun 2, 2022
Grant date	—

Abstract

Embodiments herein relate, in some examples, to an intermediate network node configured to operate in a communication network. The communication network comprises a requesting node and an executing network node comprising a computational graph model. The intermediate network node is configured with an imitation model. The imitation model is a limited version of the computational graph model, and the imitation model is a model requiring less computational resources to converge when compared to the computational graph model.

First claim

1 . An intermediate network node in a communication network that comprises a requesting node and an executing network node comprising a computational graph model, the intermediate network node comprising:
at least one processor; at least one memory connected to the at least one processor and storing an imitation model that is executed by the at least one processor to perform operations, wherein the imitation model is a limited version of the computational graph model, and wherein the imitation model is a model requiring less computational resources to converge when compared to the computational graph model.

2 . The intermediate network node according to claim 1 , wherein the imitation model comprises one or more of following compared to the computational graph model: at least one input parameter less than the computational graph model; at least one output parameter less than the computational graph model; one or more functions of less computational complexity; and at least one less internal vertex or node of the computational graph model and/or at least one less edge of the computational graph model.

3 . The intermediate network node according to claim 1 , wherein the intermediate network node builds the imitation model based on received one or more input parameters from the requesting node.

4 . The intermediate network node according to claim 3 , wherein the imitation model is built by removing one or more parts of the imitation model that have not been used within a set interval.

5 . The intermediate network node according to claim 1 , wherein the intermediate network node obtains the imitation model from the computational graph model.

6 . The intermediate network node according to claim 1 , wherein the least one memory connected to the at least one processor stores program code that is executed by the at least one processor to perform further operations comprising:
receive a request from the requesting node, wherein the request comprises one or more input parameters; and determine whether to respond to the request or to forward the one or more parameters towards the executing network node by comparing the one or more input parameters to one or more needed input parameters of the imitation model and/or based on one or more output parameters of the imitation model.

7 . The intermediate network node according to claim 6 , wherein the one or more parameters is forwarded to a second intermediate node comprising a second imitation model being a version of the computational graph model requiring less computational resources to converge when compared to the computational graph model but more computational resources than the imitation model.

8 . The intermediate network node according to claim 1 , wherein the computational graph model is a neural network and/or a decision tree.

9 . The intermediate network node according to claim 1 , wherein the intermediate network node operates between the requesting node and the executing network node.

10 . A method performed by an intermediate network node for operating in a communication network that comprises a requesting node and an executing network node comprising a computational graph model, the method comprising
obtaining an imitation model, wherein the imitation model is a limited version of the computational graph model, and wherein the imitation model is a model requiring less computational resources to converge when compared to the computational graph model.

11 . The method according to claim 10 , wherein the imitation model comprises one or more of following compared to the computational graph model: at least one input parameter less than the computational graph model; at least one output parameter less than the computational graph model; one or more functions of less computational complexity; and at least one less internal vertex or node of the computational graph model and/or at least one less edge of the computational graph model.

12 . The method according to claim 10 , wherein the imitation model is built based on received one or more input parameters from the requesting node.

13 . The method according to claim 12 , wherein the imitation model is built by removing one or more parts of the imitation model that have not been used within a set interval.

14 . The method according to claim 10 , wherein the obtain the imitation model comprises obtain the imitation model from the computational graph model.

15 . The method according to claim 10 , further comprising
receiving a request from the requesting node, wherein the request comprises one or more input parameters; and determining whether to respond to the request or to forward the one or more parameters towards the executing network node by comparing the one or more input parameters to one or more needed input parameters of the imitation model and/or based on one or more output parameters of the imitation model.

16 . The method according to claim 15 , further comprising
forwarding the one or more parameters to a second intermediate node comprising a second imitation model being a version of the computational graph model requiring less computational resources to converge when compared to the computational graph model but more computational resources than the imitation model.

17 . The method according to claim 10 , wherein the computational graph model is a neural network and/or a decision tree.

18 . The method according to claim 10 , wherein the intermediate network node is configured to operate between the requesting node and the executing network node.

19 . A computer program product comprising a non-transitory storage medium including instructions, which, when executed on at least one processor of the intermediate network node cause the the intermediate network node to perform operations comprising:
obtaining an imitation model, wherein the imitation model is a limited version of the computational graph model, and wherein the imitation model is a model requiring less computational resources to converge when compared to the computational graph model.

20 . (canceled)

21 . The computer program product of claim 19 , whereby execution of the instructions causes the intermediate network node to perform further operations comprising:
receiving a request from the requesting node, wherein the request comprises one or more input parameters; and determining whether to respond to the request or to forward the one or more parameters towards the executing network node by comparing the one or more input parameters to one or more needed input parameters of the imitation model and/or based on one or more output parameters of the imitation model.

Assignees

• Ericsson Telefon Ab L M

Inventors

• Karapantelakis Athanasios
• Andersson Markus
• Fersman Elena
• Bergkvist Adam
• Eleftheriadis Lackis
• Vandikas Konstantinos

Classifications

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/048

  Activation functions · CPC title

• G06N3/08 Primary

  Learning methods · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title