Abstract: A computer-implemented method for distributed machine learning, performed in a wireless access network including a plurality of nodes. The method includes: defining a set of neighbor relation edges which connect at least some of the nodes of the wireless access network, where each neighbor relation edge is associated with at least one neighbor relation (KPI), selecting a subset of the neighbor relation edges for each node, wherein the selected subset of neighbor relation edges is associated with one or more neighbor relation KPI that meets a pre-determined acceptance criterion, forming a relational graph for each node based on the respective selected subset of neighbor relation edges for the node, and performing distributed machine learning over the nodes in the wireless access network based on the formed relational graph for each node.

Abstract: A macromolecule membrane structure (2) comprises a membrane (3) with water-channeling integral membrane proteins (IMPS) (1) and is coated, on a first surface, with a silica layer (4). The silica layer (4) stabilizes the macromolecule membrane structure (2) and the water-channeling IMPS (1) while maintaining the water-channeling function of the water-channeling IMPs (1). As a consequence of this stabilization, the macromolecule membrane structure (2) may be used in a filtration device (5) for various filtration operations, including water purification.

Abstract: A method (100) is disclosed for managing a wireless device that is operable to connect to a communication network, wherein the communication network comprises a Radio Access Network (RAN), and wherein the wireless device has available for execution a Machine Learning (ML) model that is operable to provide an output, on the basis of which a RAN operation performed by the wireless device may be configured. The method, performed by a RAN node of the communication network, comprises, on fulfilment of a trigger condition, causing an ML model Assurance Information, MAI, Request to be sent to the wireless device (110), the MAI Request comprising an indication of the ML model to which the MAI Request relates.

Abstract: A method performed by a first network entity in a communications network is provided. The method comprises receiving a request from a second network entity, the request comprising one or more selection criteria for selecting network entities to participate in a collaborative learning process to train a model using a machine learning algorithm. The method further comprises transmitting a response message comprising an indication of whether or not the first network entity satisfies the one or more selection criteria.

Abstract: A method performed by a first node for handling aggregation of messages is disclosed. The first node determines out of a plurality of nodes, which nodes are to be paired as a pair of nodes. The pairing is to pass messages to another node for aggregation of information comprised in the messages. The information is to be masked by respective masks of opposite signs. The respective masks are to prevent the another node from accessing the information while enabling their aggregation. The determining is based on a respective measure of reliability of communications of a second node and a third node in the pair of nodes being similar, according to a criterion. The first node initiates an aggregation of the messages passed from the determined pair of nodes to the another node.

Abstract: A method performed by a first network entity in a communications network is provided. The method comprises receiving a request from a second network entity, the request comprising one or more selection criteria for selecting network entities to participate in a collaborative learning process to train a model using a machine learning algorithm. The method further comprises transmitting a response message comprising an indication of whether or not the first network entity satisfies the one or more selection criteria.

Abstract: A computer-implemented method performed by a local computing device for collaborative machine learning in a communication network is provided. The method comprises receiving from a global computing device, a plurality of global ML models. The method further comprises evaluating a metric on a set of data of the local computing device for each respective global ML model from the plurality of global ML models. The evaluating comprises (i) generating a random number, and (ii) comparing the random number to a predetermined value. The method further comprises selecting a global ML model from the plurality of global ML models, wherein the selecting is (i) a random global ML model from the plurality of global ML models when the random number is less than the predetermined value, or (ii) a global ML model from the plurality of global ML models having a greatest performance on the set of data of the local computing device when the random number is greater than the predetermined value.

Abstract: The present invention relates to a method for predicting a performance indicator for a service in a network. The method is performed by a network node of the network, and the method comprises obtaining measurement data of a metric affecting a service communicating via a radio access network, RAN, node, wherein the metric is independent of the service communicating via the RAN node, inputting the obtained measurement data into a prediction model for performance of the service communicating via the RAN node, wherein the prediction model has been trained with measurement data from the RAN node and measurement data from an end node, and predicting the performance indicator for performance of the service in the network. A network node, a computer program and a computer program product are also presented.

Abstract: In a method in a user equipment, UE, in a communications network, of determining whether to perform a handover procedure from a first network node to a second network node, a location of the UE is provided as input to a model stored on the UE, the model having been trained using a machine learning process to predict conditions on the second network node in the communications network based on the location of the UE. A prediction of conditions on the second network node at the provided location of the UE is provided by the model. The received predicted conditions are then used to determine whether to perform a handover procedure.

Abstract: A method performed by a first wireless device served by a first wireless access point in a first wireless communications network, the first wireless communications network being operated by a first network operator, comprises acquiring (202) a determination from a first reinforcement learning agent of whether to roam from the first wireless access point to a second wireless access point in a second wireless communications network, the second wireless communications network being operated by a second network operator. The method further includes roaming (204) from the first wireless access point to the second wireless access point, based on the determination.

Abstract: Some embodiments herein relate to a method performed by a wireless communication device for managing communication in a wireless communications network. The wireless communication device obtains an indicator indicating a model and one or more trained model parameters for the model, wherein the model is related to an event being one of the following events: a handover procedure, a cell reselection procedure, and a beam reselection procedure. The wireless communication device further selects the model based on the obtained indicator. The wireless communication device executes the selected model using the obtained one or more trained model parameters; and triggers a process, being associated with the event, based on an output of the executed model.

Abstract: A computer implemented method of training a reinforcement learning model for controlling a dynamic system includes generating a trajectory sample of a simulated system that corresponds to the dynamic system, the trajectory sample including a current state st of the simulated system at time t, an action at taken on the simulated system at time t according to a policy ?, a subsequent state st+1 of the simulated system following the action at, and a reward r associated with the action at, and estimating a robust target value V?(st) for the trajectory sample, wherein the robust target value V?(st) includes an expected value of a sum of the reward r and a minimum estimated value V?(st+1) of the simulated system at the subsequent state st+1 based on a plurality of transition possibilities p from the current state st in response to the action at. The method updates a value function estimator based on the robust target value, and updates the policy based on the trajectory and the value function estimator.

Abstract: Embodiments of the disclosure provide methods, apparatus and computer programs for generating a radio coverage map. A method comprises: obtaining image data of a geographical area, the image data comprising: a representation of the environment in the geographical area; and an indication of one or more transmission point locations corresponding to the locations of one or more transmission points in a wireless communications network; and applying a generative model to the image data, to generate a radio coverage map of the geographical area.

Abstract: A method performed by a first network entity in a communications network is provided. The method comprises receiving a request from a second network entity, the request comprising one or more selection criteria for selecting network entities to participate in a collaborative learning process to train a model using a machine learning algorithm. The method further comprises transmitting a response message comprising an indication of whether or not the first network entity satisfies the one or more selection criteria.

Abstract: A method performed by a co-ordination network entity in a communications network includes transmitting a request message to a network registration entity in the communications network for identification information for a plurality of candidate network entities in the communications network capable of performing collaborative learning, and receiving identification information for the plurality of candidate network entities from the network registration entity. The method further includes initiating, at one or more network entities of the plurality of candidate network entities, training of a model using a machine-learning algorithm as part of a collaborative learning process.

Abstract: A method performed by a first network entity in a communications network includes training a model to obtain a local model update including an update to values of one or more parameters of the model, in which training the model includes inputting training data into a machine learning algorithm. The method further includes applying a serialisation function to the local model update to construct a serial representation of the local model update, thereby removing information indicative of a structure of the model, and transmitting the serial representation of the local model update to an aggregator entity in the communications network.

Abstract: A method performed by a first entity in a communications network is provided. The first entity belongs to a plurality of entities configured to perform federated learning to develop a model. In the method, the first entity trains a model using a machine-learning algorithm, generating a model update. The first entity generates a first mask, receives an indication of one or more respective second masks from a subset of the remaining entities of the plurality of entities, and combines the first mask and the respective second masks to generate a combined mask. The first entity transmits an indication of the first mask to one or more third entities of the plurality of entities. The first entity applies the combined mask to the model update to generate a masked model update and transmits the masked model update to an aggregating entity of the communications network.

Abstract: A method for creating one or more machine learning models in a model training engine is provided. The method includes providing to a user, via a graphical user interface, a selection of components for a machine learning model, at least one component having a computational layer including one or more coefficients associated with a component attribute. The method also includes displaying, in the graphical user interface, a component selected by the user, including a selected value of the component attribute and executing the machine learning model with a training archive as an input, to obtain an output indicative of a desired feature of the training archive. The method also includes comparing the output with a desirable feature value, and modifying at least one coefficient in the component of the machine learning model based on a difference between the output from the machine learning model and the desirable feature value.

Abstract: The present invention relates to a method for predicting a performance indicator for a service in a network. The method is performed by a network node of the network, and the method comprises obtaining measurement data of a metric affecting a service communicating via a radio access network, RAN, node, wherein the metric is independent of the service communicating via the RAN node, inputting the obtained measurement data into a prediction model for performance of the service communicating via the RAN node, wherein the prediction model has been trained with measurement data from the RAN node and measurement data from an end node, and predicting the performance indicator for performance of the service in the network. A network node, a computer program and a computer program product are also presented.

Abstract: The present invention relates to a method for predicting a performance indicator for a service in a network. The method is performed by a network node of the network, and the method comprises obtaining measurement data of a metric affecting a service communicating via a radio access network, RAN, node, wherein the metric is independent of the service communicating via the RAN node, inputting the obtained measurement data into a prediction model for performance of the service communicating via the RAN node, wherein the prediction model has been trained with measurement data from the RAN node and measurement data from an end node, and predicting the performance indicator for performance of the service in the network. A network node, a computer program and a computer program product are also presented.