Method for identifying potential machine learning model candidates to collaborate in telecom networks

A network device for executing a collaboration client for identifying cells of a mobile network for machine learning collaboration, the mobile network having a plurality of cells, wherein each cell of the plurality of cells is associated with different edge nodes of the mobile network and with a respective machine learning model for each cell, the network device comprising:

• a non-transitory computer-readable medium having stored therein a collaboration client;
• and
• a processor coupled to the non-transitory computer-readable medium, the processor to execute the collaboration client, the collaboration client to receive clustering information from a collaboration server, which collaboration server collects features for the plurality of cells to generate at least one feature vector for each of the plurality of cells, wherein the features are extracted at the respective machine learning model at each edge node to generate the clustering information, the collaboration client further to perform a collaboration pre-check with each cluster cell of a cluster identified in the clustering information, wherein pre-check information of a pair of cells of the cluster is used to determine performance related metrics used to determine whether the pair of cells can collaborate, to send pre-check results to the collaboration server, to receive collaboration information from the collaboration server, and to exchange information with one or more cells identified in the collaboration information, the exchanged information for machine learning models utilized by the network device and the one or more cells.

An electronic device to execute a plurality of virtual machines, the plurality of virtual machines to execute a method of a collaboration server for identifying cells of a mobile network for machine learning collaboration, the mobile network having a plurality of cells, wherein each cell of the plurality of cells is associated with different edge nodes of the mobile network and with a respective machine learning model for each cell, the electronic device comprising:

• a non-transitory computer-readable medium having stored therein a collaboration server;
• and
• a processor to execute the plurality of virtual machines, one of the plurality of virtual machines to execute the collaboration server, the collaboration server to manage collection of features for the plurality of cells to generate at least one feature vector for each of the plurality of cells, wherein the features are extracted at the respective machine learning model at each edge node, to determine a cluster of cells within the plurality of cells based on similarity in feature vectors between at least two cells in the plurality of cells, to send cluster information to each cell of the cluster, to receive cluster pre-check information from each cell of the cluster, wherein pre-check information of a pair of cells of the cluster is used to determine performance related metrics used to determine whether the pair of cells can collaborate, and to determine, based on the received pre-check information, at least a first cell and a second cell in the cluster to exchange information for machine learning model utilized by the first cell and machine learning model utilized by the second cell.

A computing device to implement a control plane of a software defined networking network, the computing device to execute a method of a collaboration server for identifying cells of a mobile network for machine learning collaboration, the mobile network having a plurality of cells, wherein each cell of the plurality of cells is associated with different edge nodes of the mobile network and with a respective machine learning model for each cell, the computing device comprising:

• a non-transitory computer readable medium having stored therein a collaboration server;
• and
• a processor coupled to the non-transitory computer readable medium, the processor to execute the collaboration server, the collaboration server to manage collection of features for the plurality of cells to generate at least one feature vector for each of the plurality of cells, wherein the features are extracted at the respective machine learning model at each edge node, to determine a cluster of cells within the plurality of cells based on similarity in feature vectors between at least two cells in the plurality of cells, to send cluster information to each cell of the cluster, to receive cluster pre-check information from each cell of the cluster, wherein pre-check information of a pair of cells of the cluster is used to determine performance related metrics used to determine whether the pair of cells can collaborate, and to determine, based on the received cluster pre-check information, at least a first cell and a second cell in the cluster to exchange information for machine learning model utilized by the first cell and machine learning model utilized by the second cell.