Abstract: There is provided an apparatus comprising: at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to receive, at an analytics producer, a request for analytics, from an analytics consumer, wherein the request comprises a first time period, TA, determine a second time period, DS for an expected time to provide the requested analytics at the analytics producer, provide the analytics based on the request in a third time period, TD, determine, based on TA, DS and TD, which of at least one of a plurality of parameters is met and increment a counter for the determined parameter.

Abstract: Systems, methods, apparatuses, and computer program products for cluster based local ML model training host selection in asynchronous federated learning model collection. A method may include transmitting local model parameters following training of a model of at least one network element to a network node to be used to cluster the at least one network element with one or more other network elements. The method may also include training, by one or more network elements selected from the cluster, an aggregated model using the local model parameters. The method may further include transmitting, by the one or more network elements selected from the cluster, updated local model parameters of the at least one network element as a result of the training to the network node.

Abstract: Cognitive functions in a network, each comprising a learning module adapted to observe characteristics of the network and an inference module adapted to conclude, from the characteristics observed by the learning module, on a behavior of the network resulting from the observed characteristics, are controlled by halting and/or restarting operation of the learning module of a cognitive function of the cognitive functions, where operation of the inference module of the cognitive function is controlled independently from the halting and/or restarting the operation of the learning module of the cognitive function.

Abstract: Apparatuses and methods in a communication system are disclosed. In a network element, an encoder module obtains as an input network data that is representative of the current condition of the communications network, the network data comprising a plurality of values indicative of the performance of network elements and performs (800) feature reduction providing at its output a set of activations. A clustering module performs (802) batch normalisation and an amplitude limitation to the output of the encoder module to obtain normalised activations. A clustering control module calculates a projection of the normalised activations and determines (804) a clustering loss. A decoder module calculates (806) a reconstruction loss. The network element backpropagates the reconstruction loss and the clustering loss through the modules.

Abstract: First user equipments are detected out of a plurality of user equipments of a cellular communication system (S201). The user equipments respectively correspond to a distributed node of a federated machine-learning concept and respectively generate a partial machine-learning model, wherein partial machine-learning models generated by the plurality of user equipments are to be used to update a global machine-learning model at the network side of the cellular communication system. The first user equipments are user equipments comprising ready partial machine-learning models.

Abstract: Systems, methods, apparatuses, and computer program products for supporting cognitive service driven handover are provided. One method may include signaling at least one user equipment to start performing and logging quality of experience (QoE) related measurements, and transmitting a request for a report on handover performance to the at least one user equipment. The method may also include receiving the handover performance report, from the at least one user equipment, including at least one service identifier of a service running on the at least one user equipment and the quality of experience (QoE) related measurements.

Abstract: There is provided an apparatus arranged to: receive, from a configuration controller, a request for feedback on a network configuration in a communications network; request feedback on the network configuration from a plurality of network critic functions; aggregate feedback received from at least part of the plurality of network critic functions in response to the request; and provide the aggregated feedback to the configuration controller.

Abstract: According to an example aspect of the present invention, there is provided an apparatus comprising at least one processing core, at least one memory including computer program code, the at least one memory and the computer program code being configured to, with the at least one processing core, cause the apparatus at least to generate, using a machine learning mechanism, a set of test instructions, wherein each one of the test instructions is configured to trigger a first network to change one or more technical characteristics of the first network, the test instructions not comprising configuration changes in nodes of the first network that are usable in changing the one or more technical characteristics of the first network, and use a block of text as input in the generation of the set of test instructions.

Abstract: Systems, methods, apparatuses, and computer program products for managing machine learning training. One method may include a machine learning training function receiving a request to instantiate a machine learning training job from a consumer, and instantiating the requested machine learning training job. The machine learning training function may transmit a notification to the consumer indicating that the machine learning training job has been instantiated.

Abstract: It is provided a method comprising: monitoring if a distributed training host receives a request from a meta-training host to provide a machine learning model; checking whether a link from the distributed training host to the meta-training host is required for another data communication having higher priority than providing the machine learning model such that the other data communication will block the link for the providing the machine learning model to the meta-training host; informing the meta-training host, in response to the request, that the link is required for the other data communication if the link is required for the other data communication.

Abstract: This document discloses a solution for performing exploration of radio resource management actions. According to an aspect, a method in a terminal device includes: receiving configuration information from a network node of a radio access network; entering, in response to the configuration information and in a state of not having a need to transfer data, an exploration mode where a reduced set of radio connection functions are enabled compared with a default operating mode; triggering, in the exploration mode, an explorative handover from a source cell managed by the network node to a target cell; establishing a radio connection with a target network node as a result of the explorative handover to the target cell managed by the target network node; acquiring measurement data of the radio connection and transmitting the measurement data to the target network node.

Abstract: It is provided a method comprising: monitoring if a request to train a machine learning sub-model is received from a meta-training host; generating training data; training the machine learning sub-model by at least a first subset of the training data if the request is received and at least the first subset of the training data is generated; checking if a predefined condition related to the machine learning sub-model is fulfilled; providing the trained machine learning sub-model and at least a second subset of the training data to the meta-training host if the condition is fulfilled.

Abstract: It is provided a method, comprising monitoring if a generic objective for a network is received; translating the generic objective into specific objectives based on a behavioral matrix if the generic objective is received, wherein each of the specific objectives is specific for a respective network element; requesting, for each of the specific objectives, an automation function of the respective network element to achieve the specific objective, identifying, for each of the specific objectives, based on a stored association table, a distributed control function controlling the automation function of the respective network element; informing, for each of the specific objectives, the identified distributed control function on the specific objective for the respective network element; supervising if a feedback is received from one of the distributed control functions, wherein the feedback indicates to which degree one of the specific objectives is achieved; adapting the behavioral matrix based on the feedback.

Abstract: There is provided the apparatus comprising: at least one processor; and at least one memory comprising computer code that, when executed by the at least one processor, causes the apparatus to: identify a potential problem in a network comprising at least one network automation function; signal an indication of said potential problem to at least one network automation function of said network and a request for a proposal to address said problem; receive at least one proposal in response to said signalling; determine policy changes for addressing said potential problem in dependence on said at least one proposal; and implement said policy changes.

Abstract: The disclosure relates to an apparatus comprising means for: receiving a request from an operator to instantiate a closed loop for managing a communication system; identifying closed loop components based on the request; and instantiating a closed loop for managing a communication system with the closed looped components.

Abstract: There are provided measures for configuration of power saving groups.

Abstract: Systems, methods, apparatuses, and computer program products for a service-centric mobility-based traffic steering.

Abstract: Systems, methods, apparatuses, and computer program products for automating context-specific network function configuration are provided. One method may include receiving an objective model including rules defining key performance indicator (KPI) targets and relative prioritizations of the KPI targets for a communications system. The method may also include automatically determining, using the received objective model, a context model including at least a description of properties of one or more contexts, and generating or selecting at least one of function configuration parameter values (FCVs) or network configuration parameter values (NCPs) according to the context model.

Abstract: To configure network automation functions, a scope administration function assigns, per a region in an operator-defined scope space, the region to one or more network automation function amongst a plurality of coexisting network automation functions. Then, per a network automation function assigned at least to one region, the network automation function is configured to control the at least one region assigned to the network automation function.

Abstract: Cognitive functions in a network, each comprising a learning module adapted to observe characteristics of the network and an inference module adapted to conclude, from the characteristics observed by the learning module, on a behavior of the network resulting from the observed characteristics, are controlled by halting and/or restarting operation of the learning module of a cognitive function of the cognitive functions, where operation of the inference module of the cognitive function is controlled independently from the halting and/or restarting the operation of the learning module of the cognitive function.