Abstract: A computer-implemented method is provided performed by a client computing device for decentralized learning based on local learning at the client computing device is provided. The method includes training a local M, model based on an activation function using a local parameter set and a reference parameter set to obtain a setting for respective 5 local parameters in the local parameter set that minimizes a training loss wherein the activation function preserves agreements and discourages disagreements between the local parameter set and the reference parameter set. The method further includes sending the trained local ML model to a server computing device. The method further includes receiving, from the server computing device, a global ML model that meets a convergence criterion. A 10 method performed by a server computing device, and related methods and apparatuses are also provided.

Abstract: There is provided a method for estimating power saved in a first network. The method is performed by a system. The method includes estimating the power saved in the first network as a difference between: a relationship between power consumption in the first network as a function of physical resource block usage in the first network when a power saving feature is deactivated in the first network, and at least one data point indicative of the power consumption in the first network for a physical resource block usage value when the power saving feature is activated in the first network.

Abstract: A system (200), a first network node (240), a method, a computer program and a computer program product for training of a Federated Learning. FL, model is disclosed. The system comprises network nodes. One of the network nodes is a first network node. Each network node has access to a part of the network data. The system obtains network information and determines groups of network nodes and assigns each network node to one of the determined groups based on the network information, each determined group of network nodes comprising at least two network nodes. For each of the groups, the system appoints a second network node from among the at least two network nodes, informs the at least two network nodes about the appointed second network node and trains an FL model using the parts of the network data accessible by the at least two network nodes.

Abstract: A computer-implemented method is provided performed by a network node (600, 1200) including a first machine learning, ML, model (105) representing an enhanced policy to optimize a radio access network, RAN. The method includes receiving (901) an initial policy. The method further includes training (903) a second ML model based on a plurality of interactions with an environment including a portion of the RAN. The method further includes training (909) the first ML learning model to learn the enhanced policy with data from the trained second ML model and the environment; and deploying (911) the trained first ML model including the enhanced policy to optimize the RAN.

Abstract: A method performed by a vehicle manager (150) for managing a remote-controlled vehicle (160). The vehicle manager (150) obtains an original route to the destination passing a sequence of radio access points in a wireless network that a wireless device in the vehicle (160) will connect to in a communication with the vehicle manager (150) when the vehicle (160) travels along the original route. The vehicle manager (150) detects at least one deficient radio access point of the original route. The vehicle manager (150) then instructs the vehicle (160) to adapt its behaviour based on said detecting.

Abstract: Methods and systems for source machine learning (ML) model selection for the transfer learning. A method may include receiving a source ML model request from a target domain, determining candidate source ML models, calculating a model quality score for each of the candidate source ML models, using the calculated model quality scores to select candidate source ML models, sending the selected candidate source ML models to the target domain, receiving fine-tuned ML model weights for fine-tuned ML models, and calculating a model quality score for each of the fine-tuned ML models. The method may include determining, for each of the fine-tuned ML models, a ranking and/or a deployment recommendation for the fine-tuned ML model based on the model quality score for the fine-tuned ML model and sending, for each of the fine-tuned ML models, the ranking and/or the deployment recommendation for the fine-tuned ML model to the target domain.

Abstract: Methods and leader computing devices for developing machine-learning models. A method comprises receiving, at a leader computing device from each of a plurality of worker computing devices, weights and model architecture information for part of a trained ML model. The method further comprises determining, at the leader computing device, a common portion of the parts of trained ML models that is useable by all of the plurality of worker computing devices, and generating, at the leader computing device, an updated common portion of the ML model using the common portion of the parts of trained ML models and the weights and model architecture information from each of the plurality of worker computing devices. The method further comprises initiating transmission of the updated common portion of the ML model to the worker computing devices.

Abstract: Embodiments herein disclose a method for selecting a machine learning model to be deployed in an execution environment having resource constraints. The method comprises receiving, by an apparatus, a request for a machine learning model solving a task T using a feature set F. Further, the method includes retrieving, from a model store, a first set of machine learning models that solves the task T using at least a subset of features F. The complexity of each machine learning model in the first set of machine learning models is calculated. The method includes determining, from the first set of machine learning models, at least one suitable machine learning model to be deployed, wherein the determining is based on the calculated complexity and the resource constraints of the execution environment.

Abstract: Embodiments herein disclose a method for selecting a machine learning model to be deployed in an execution environment having resource constraints. The method comprises receiving, by an apparatus, a request for a machine learning model solving a task T using a feature set F. Further, the method includes retrieving, from a model store, a first set of machine learning models that solves the task T using at least a subset of features F. The complexity of each machine learning model in the first set of machine learning models is calculated. The method includes determining, from the first set of machine learning models, at least one suitable machine learning model to be deployed, wherein the determining is based on the calculated complexity and the resource constraints of the execution environment.

Abstract: A computer-implemented method is provided for determining whether a machine learning model in a candidate source domain is suitable for use in a target domain, wherein the machine learning model is trained with one or more features. The method comprises for each feature: determining one or more target measurement configurations indicating how data for the feature can be generated in the target domain; for each feature, performing, for each of a plurality of candidate source domains, the steps of: determining one or more source measurement configurations indicating how data for the feature can be generated in the candidate source domain, determining a similarity metric indicative of a similarity between the one or more source measurement configurations and the one or more target measurement configurations; and based on the similarity metrics determined for each feature for the plurality of candidate source domains, selecting one or more selected source domains from the plurality of candidate source domains.

Abstract: A method by a client computing device having one or more sensors for collecting data, includes obtaining information identifying a first set of measurable features, each feature being associated with a measurement specification. For each feature of the first set of measurable features, determining whether there is at least one sensor of the one or more sensors satisfying the associated measurement specification. If there is at least one sensor of the one or more sensors satisfying the associated measurement specification, estimating a resource usage. Determining a first subset of the first set of measurable features and sending information identifying the first subset of the first set of measurable features. When the client computing device is determined to belong to the first group of computing devices, performing training of the machine learning model using the first group of computing devices.

Abstract: There is provided a method for estimating power saved in a first network. The method is performed by a system. The method includes estimating the power saved in the first network as a difference between: a relationship between power consumption in the first network as a function of physical resource block usage in the first network when a power saving feature is deactivated in the first network, and at least one data point indicative of the power consumption in the first network for a physical resource block usage value when the power saving feature is activated in the first network.

Abstract: A method for machine-learning adaptation comprises identifying (110) a plurality of machine-learning source domain candidates and calculating (120), for each of the identified machine-learning source domain candidates, a diversity metric, where the diversity metric represents a marginalized measure of sample diversity of the respective machine-learning source domain candidate. The method further comprises selecting (130) the identified machine-learning source domain candidates having a highest diversity metric among the calculated diversity metrics and applying (140) the selected machine-learning source domain candidate to a target domain in a new or changed execution environment. The diversity metric may be calculated based on information theoretic measures, for example, such as based on a one-parameter measure of generalized entropy.

Abstract: In a communication system, a first network node is configured to execute at least one service application executing a first service and at least one analytics application executing at least part of a distributed analytics service. The first network node obtains information about a new telecommunication service and transmits, to a second network node in the communication system, a request for a policy for the new telecommunication service. The first network node receives, from the second network node, the policy for the new telecommunication service and updates a currently applied policy on the basis of the received policy. The updated policy rebalances resources allocated from a shared computing resource pool of the first network node between the new telecommunication service and the at least one analytics application such that the new telecommunication service maintains adherence to the one or more requirements of a service level agreement.

Abstract: A master node, a local node, a service assurance system, and a respective method performed thereby for predicting one or more metrics associated with a communication network are provided. The method performed by the master node operable in the communication network comprises receiving prediction(s) based on training data from local nodes in the communication network; and determining weight parameter(s) associated with the local nodes based on the current received prediction(s) and past received predictions. The method further comprises adjusting a respective local reporting policy for one or more local nodes based on the determined weight parameter(s).

Abstract: A method and a network agent for providing cell assignment for a wireless device served by a network node. An input vector is created for a set of candidate cells based on measurements by the wireless device and/or by the network node related to performance and signals. A future effect of assigning the wireless device to a candidate cell is estimated for each candidate cell by applying the created input vector to an effect estimation function which may be a Q-learning function. A cell in the set of candidate cells is then determined and assigned for the wireless device, based on the estimated future effects of the candidate cells. The cell that provides the best future effect may be selected for cell assignment.

Abstract: In a communication system, a first network node is configured to execute at least one service application executing a first service and at least one analytics application executing at least part of a distributed analytics service. The first network node obtains information about a new telecommunication service and transmits, to a second network node in the communication system, a request for a policy for the new telecommunication service. The first network node receives, from the second network node, the policy for the new telecommunication service and updates a currently applied policy on the basis of the received policy. The updated policy rebalances resources allocated from a shared computing resource pool of the first network node between the new telecommunication service and the at least one analytics application such that the new telecommunication service maintains adherence to the one or more requirements of a service level agreement.

Abstract: It is disclosed a network node (200, 220, 300, 400, 500, 600, 80) and a method for executing an analytics task. The network node is adapted to reside in a telecommunication network, and to be supported by a distributed cloud infrastructure. The network node is adapted to execute at least one service application (202) executing a first service and at least one analytics application (204) executing at least part of a distributed analytics service, where the network node further is adapted to comprise a node policy agent (206, 226, 304, 504, 606) and a node manager (210, 230, 306, 406, 506, 608). By applying a policy restrictions on the analytics task executed on the same network node as a service application, service level agreements for the service application in a telecommunication cloud can be upheld.

Abstract: A method for predicting a location of a fault in a system is described. The method includes obtaining a faulty unit pertaining to a Supervised Self-Organizing Map (SSOM), wherein the faulty unit has been derived from a sample of machine-level metrics for the system and the SSOM has been trained with a first layer of machine-level metrics and a second layer of service-level metrics. The method further includes expanding a circle originating at the faulty unit until a number of normal units of the SSOM falls within the circle, computing, for each machine-level metric, a score based on differences between the faulty and normal units, wherein a set of scores comprises the respective score for each machine-level metric, and selecting a sub-set of the set, wherein the sub-set comprises the greatest scores of the set, whereby the fault is predicted as located according to metrics represented by the sub-set.

Abstract: A method and a machine learning manager (100) for handling prediction of service characteristics using machine learning applied in a target domain (102B). A source model MS used for machine learning pre-trained in a source domain (102A) is obtained, and a transfer configuration that divides the source model into a fixed first part and a non-fixed second part is selected. A target model is created by applying the selected transfer configuration on the source model so that the target model is divided into said first and second parts. The second part is then trained using observations collected in the target domain, and the target model MT with the first part and the trained second part is provided for prediction of service characteristics in the target domain.