Abstract: In one aspect there is a method for detecting whether a first system should be declared as an impostor. In one embodiment, the method includes generating baseline system output information using first system input; exposing the first system to the first system input; obtaining first system output information generated by the first system based on the first system's exposure to the first system input; determining a first similarity measure that indicates a similarity between the baseline system output information and the first system output information generated by the first system; and using the first similarity measure, determining whether or not to declare the first system as an impostor.

Abstract: A computer implemented method for translating a sensory input to a sensory output for communication between first and second entities. The first entity includes a user and the second entity includes a user or a computing system. The method includes obtaining a first profile specifying a communication capability of the first entity, and a second profile specifying a communication capability of the second entity and obtaining a trained Machine Learning (ML) model operable to map an input sensory communication in accordance with one of the first or second profiles to an output sensory communication in accordance with the other of the profiles. The method further includes receiving an input sensory communication from one of the first or second entities, using the ML model to map the input sensory communication to an output sensory communication, and providing the output sensory communication to the other of the entities.

Abstract: The present application relates to a method for verifying a Channel Quality Indicator (CQI) measurement performed by a first wireless device in a communication network. The communication network includes a Radio Access Network (RAN). The method includes: obtaining a first CQI measurement report including a first CQI measurement value of a measurement performed by the first wireless device on for a first downlink channel between a first RAN node and the first wireless device; identifying at least one neighbour wireless device to the first wireless device; obtaining at least one neighbour CQI measurement report from the at least one neighbour wireless device; aggregating the at least one neighbour CQI measurement report into a reference CQI measurement report set; and verifying the first CQI measurement value using the reference CQI measurement report set. The present application also relates to a verification node and a computer program product.

Abstract: The disclosure relates to a computer-implemented method for configuring at least one operation parameter in a network cell (140) served by a network node (110) of a communication network (100), the method comprising: (S210) obtaining at least one cell feature for each one of a plurality of network cells (140), the cell feature representing a cell state of respective network cell (140), (S220) clustering a number of network cells (140) into a number of cell clusters (160), wherein the clustering of network cells (140) is performed based on the obtained at least one cell feature of respective network cell (140), (S230) determining one network cell (140) of respective cell cluster (160) to be a representor cell (170), and for at least one cell cluster (160): (S240) determining at least one action (a) to be performed for configuring at least one operation parameter for a number of network cells (140) of respective cell cluster (160).

Abstract: There is provided a method for consolidating explanations associated with actions proposed based on a current state of a system and an intent. The method includes acquiring first and second explanations, the first and second explanations being associated with a proposed action or with different actions, wherein each of the first and second explanations includes one or more constraints, combining constraints from the first and second explanations to form a set of constraints D, generating a planning problem P=<K, A, I, G, Cost>, wherein K consists of a set of predicates F and the set of constraints D, wherein A represents a set of possible actions, I represents an initial state of the system, G represents a goal state of the system, and Cost represents cost values associated with each constraint, and determining a solution for the planning problem.

Abstract: Methods and apparatus for implementing reinforcement learning (RL) are provided. A method of operation for a node implementing RL, wherein the node instructs actions in an environment in accordance with a policy generated by a RL agent, wherein the RL agent models the environment and encodes a state of the environment using a set of features, comprises obtaining an intent, wherein the intent specifies one or more criteria to be satisfied by the environment. The method further comprises determining a Companion Markov Decision Process (CMDP) that encodes states of the environment using a subset of the set of features used by the RL agent. The method further comprises generating a finite state automaton that represents the intent as a series of logic states, and computing a product of CMDP output states and logic states, wherein the product contains all of the potential combinations of a CMDP output state and a logic state.

Abstract: Methods and apparatus for addressing intents using machine learning (ML) are provided. A method of operation for a node implementing ML, wherein the node instructs actions in an environment in accordance with a policy generated by a ML agent, and wherein the ML agent models the environment, includes obtaining an intent, wherein the intent specifies one or more criteria to be satisfied by the environment. The method further includes determining an intent cluster from among a plurality of intent clusters to which the intent maps, the determination being based on the criteria specified by the intent, and setting initialisation parameters for a ML model to be used to model the intent, based on the determined intent cluster. The method also includes training the ML model using training data specific to the intent, and generating one or more suggested actions to be performed on the environment using the trained ML model.

Abstract: A computer implemented method for determining, in coordination with a second device, whether a first device can perform a first action. The method comprises sending (202) a first message to the second device, wherein the first message comprises i) an indication of the first action to be performed by the first device and ii) a first time interval indicating when the first action is to be performed. the method further comprises receiving (204) a second message from the second device according to an argumentation protocol defined between the first device and the second device, wherein the second message comprises i) a first response related to whether the first device can perform the first action, and ii) a second time interval in which the first response is applicable. The first response is valid according to the argumentation protocol if the second time interval overlaps the first time interval.

Abstract: Computer implemented methods are disclosed for managing, and for facilitating management of an environment within a domain, the environment being operable to perform a task. The management method includes for each of a plurality of possible actions that may be executed on the environment in a current state of the environment, obtaining a risk contour and predicted reward associated with the possible action. The method further includes selecting, from among the possible actions and on the basis of the risk contours and predicted reward, an action for execution on the environment, and initiating execution of the selected action. A risk contour associated with a possible action includes probabilities of the environment entering possible future states on execution of the action, confidence values associated with the probabilities, and for the possible future states, a representation of a domain level consequence of the environment entering the possible future state.

Abstract: A method for operating a telecom network having an SLA agreement is disclosed. The method comprises obtaining annotated alarm data comprising an indication of a threshold crossing of at least one state variable of the telecom network defined by the SLA. The annotated alarm data further comprises an indication of a desired value of the at least one state variable of the telecom network. Moreover, the method comprises determining a desired state of the telecom network based on the desired value of at least one state variable, and selecting a set of policy actions from a policy action bank. The policy action bank comprises a plurality of policy actions, where each policy action is associated with at least one estimated action effect.

Abstract: Methods provide wireless communication using a plurality of Antenna Processing Units APUs distributed along a radio stripe and sharing a bus along the radio stripe. Access is provided to a plurality of APU activation/deactivation states for a respective plurality of environmental conditions, wherein each one of the plurality of APU activation/deactivation states defines APUs of the plurality of APUs that are activated and APUs of the plurality of APUs that are deactivated for the respective one of the plurality of environmental conditions. Responsive to detecting a first one of the plurality of environmental conditions, a first one of the plurality of APU activation/deactivation states corresponding to the first one of the plurality of environmental conditions is applied to activate a first subset of the APUs and to deactivate a second subset of the APUs, wherein the first and second subsets of APUs are mutually exclusive.

Abstract: A computer implemented method performed by a node in a communications network for scheduling transmissions of a plurality of Internet of Things (IoT) devices to network resources in the communications network includes obtaining transmission patterns for the IoT devices. The method then includes clustering the IoT devices into clusters based on the obtained transmission patterns. The method then includes scheduling transmissions of IoT devices in different clusters to different network resources, thereby increasing synchronisation of the transmissions scheduled on each network resource and allowing for increased periods of inactivity of the network resources between transmissions.

Abstract: Embodiments described herein relate to methods and apparatuses for training a neural network. A method comprises receiving an input data set at a layer of the neural network; performing a forward pass and a backward pass on the input data set to determine regular output data; calculating a first loss associated with the regular output data; performing a quantized forward pass and a quantized backward pass on the input data set to determine quantized output data; calculating a second loss associated with the quantized output data; comparing the first loss to the second loss; and based on the comparison determining whether to reduce the input data set to provide a reduced data set.

Abstract: Embodiments herein relate to a method performed by a first network node (111) configured to operate in a communication network, wherein the first network node (111) is configured with a first machine learning, ML, model and being connected to a second network node (112) configured with a second ML model associated with the first ML model. The first network node obtains first data, from the second network node (112), associated with the second ML model; and upon a condition being fulfilled relating to the obtained first data, the first network node (111) keeps the first ML model as is, following a first evolution branch, and creates an updated version of the first ML model, following a second evolution branch, taking the obtained first data into account.

Abstract: A method for operating a telecom network having an SLA agreement is disclosed. The method comprises obtaining annotated alarm data comprising an indication of a threshold crossing of at least one state variable of the telecom network defined by the SLA. The annotated alarm data further comprises an indication of a desired value of the at least one state variable of the telecom network. Moreover, the method comprises determining a desired state of the telecom network based on the desired value of at least one state variable, and selecting a set of policy actions from a policy action bank. The policy action bank comprises a plurality of policy actions, where each policy action is associated with at least one estimated action effect.

Abstract: A first network node operating in a communication node can receive a plan indicating tasks to be performed by at least one second network node operating in the communication network. The first network node can determine a network slice schedule based on the tasks to be performed by the at least one second network node operating in the communication network. The first network node can provide the network slice schedule to a network slice orchestrator operating in the communication network.

Abstract: In one aspect there is a method for detecting whether a first system should be declared as an impostor. In one embodiment, the method includes generating baseline system output information using first system input; exposing the first system to the first system input; obtaining first system output information generated by the first system based on the first system's exposure to the first system input; determining a first similarity measure that indicates a similarity between the baseline system output information and the first system output information generated by the first system; and using the first similarity measure, determining whether or not to declare the first system as an impostor.

Abstract: Methods provide wireless communication using a plurality of Antenna Processing Units APUs distributed along a radio stripe and sharing a bus along the radio stripe. Access is provided to a plurality of APU activation/deactivation states for a respective plurality of environmental conditions, wherein each one of the plurality of APU activation/deactivation states defines APUs of the plurality of APUs that are activated and APUs of the plurality of APUs that are deactivated for the respective one of the plurality of environmental conditions. Responsive to detecting a first one of the plurality of environmental conditions, a first one of the plurality of APU activation/deactivation states corresponding to the first one of the plurality of environmental conditions is applied to activate a first subset of the APUs and to deactivate a second subset of the APUs, wherein the first and second subsets of APUs are mutually exclusive.

Abstract: A mobile device for travelling to a destination using a communication network is provided. The mobile device obtains from one or more network nodes in the communication network, a first path over a first sub-region of a region, comprising one or more sub-regions, towards the destination. The mobile device further follows the first path; and upon entering, or being in, a second sub-region of the region having a level of connectivity below a threshold, switches to follow a second path calculated locally at the mobile device.

Abstract: A method for configuring devices for a user is provided. The method includes creating a device-set profile for the user. The device-set profile includes a first set of configuration parameters for a first set of devices at a first location. The device-set profile further includes usage characteristics information and device interactions information for devices in the first set of devices. The method further includes retrieving the device-set profile for the user; discovering a second set of devices at a second location; and automatically configuring the second set of devices based on the device-set profile for the user.