A method in a telecommunications system including a Data Network, DN, a base station, a connection via the base station from the DN to a User Equipment, UE, executing a UE application producing application data, and an algorithm entity on the DN, wherein the base station transmits network configuration information to a DN application executing on the algorithm entity, the DN application produces and transmits a filtering configuration based on the network configuration information to the UE for use in filtering the application data before transmission to the algorithm entity, allowing the UE to produce application data filtered according to the filtering configuration, and the connection transmits the filtered application data to the algorithm entity.

A method in a telecommunications system including a Data Network, DN, a base station, a connection via the base station from the DN to a User Equipment, UE, executing a UE application producing application data, and an algorithm entity on the DN, wherein the base station transmits network configuration information to a DN application executing on the algorithm entity, the DN application produces and transmits a filtering configuration based on the network configuration information to the UE for use in filtering the application data before transmission to the algorithm entity, allowing the UE to produce application data filtered according to the filtering configuration, and the connection transmits the filtered application data to the algorithm entity.

Diagnosis is assisted by acquiring microscopical observation image data while specifying the position, classifying the image data into histological types with the use of AI, and reconstructing the classification result in a whole lesion. There is provided a pathological diagnosis assisting method that can provide an assistance technology which performs a pathological diagnosis efficiently with satisfactory accuracy by HE staining which is usually used by pathologists. Furthermore, there are provided a pathological diagnosis assisting system, a pathological diagnosis assisting program, and a pre-trained model.

Disclosed are a semantic image extrapolation method and a semantic image extrapolation apparatus. The present invention provides a technique for generating an empty region for image-extension in an image by using an extrapolated segmentation map and an inpainting technique. The present invention is to provide, considering that there is no information in an empty region for image-extension in an image, a semantic image extrapolation method, of first generating an extrapolated segmentation map on the basis of a segmentation map from an input image, and filling the empty region for image-extension in the image with information on the basis of the extrapolated segmentation map and the input image.

The possibility of a work machine 40 being affected by a disaster in a second designated area including an existence position of the work machine 40 is predicted based on an amount of rainfall in a first designated area. A hazard map representing a result of the prediction of the possibility of the work machine 40 being affected by a disaster in the second designated area is outputted to a remote output interface 220 in a remote operation apparatus 20 (a client) (or a management output interface 620 in a management client 60). Accordingly, a user can take measures to reduce the possibility of the work machine being affected by a disaster, for example, to communicate with the persons involved in order to move the work machine 40 from a current position.

The possibility of a work machine 40 being affected by a disaster in a second designated area including an existence position of the work machine 40 is predicted based on an amount of rainfall in a first designated area. A hazard map representing a result of the prediction of the possibility of the work machine 40 being affected by a disaster in the second designated area is outputted to a remote output interface 220 in a remote operation apparatus 20 (a client) (or a management output interface 620 in a management client 60). Accordingly, a user can take measures to reduce the possibility of the work machine being affected by a disaster, for example, to communicate with the persons involved in order to move the work machine 40 from a current position.

An initially trained machine learning model is used by an active learning module to generate candidate triples, which are fed into an expert system for verification. As a result, the expert system outputs novel facts that are used for retraining the machine learning model. This approach consolidates expert systems with machine learning through iterations of an active learning loop, by bringing the two paradigms together, which is in general difficult because training of a neural network (machine learning) requires differentiable functions and rules (used by expert systems) tend not to be differentiable. The method and system provide a data augmentation strategy where the expert system acts as an oracle and outputs the novel facts, which provide labels for the candidate triples. The novel facts provide critical information from the oracle that is injected into the machine learning model at the retraining stage, thus allowing to increase its generalization performance.

An initially trained machine learning model is used by an active learning module to generate candidate triples, which are fed into an expert system for verification. As a result, the expert system outputs novel facts that are used for retraining the machine learning model. This approach consolidates expert systems with machine learning through iterations of an active learning loop, by bringing the two paradigms together, which is in general difficult because training of a neural network (machine learning) requires differentiable functions and rules (used by expert systems) tend not to be differentiable. The method and system provide a data augmentation strategy where the expert system acts as an oracle and outputs the novel facts, which provide labels for the candidate triples. The novel facts provide critical information from the oracle that is injected into the machine learning model at the retraining stage, thus allowing to increase its generalization performance.

Example aspects include techniques for accurate and expeditious cloud service health communication to customers. These techniques may include determining that a service health incident has customer impact, the service health incident corresponding to an outage of one or more services of a cloud computing platform, identifying a plurality of customers impacted by the service health incident, and predicting, based on the service health incident and one or more other service health incidents, aggregated incident information identifying a plurality of service health incidents associated with the outage of the one or more services. In addition, the techniques may include identifying the one or more services associated with the service health incident, and transmitting, based at least in part on the aggregated incident information and the one or more services, a health notification to the plurality of customers.

Example aspects include techniques for accurate and expeditious cloud service health communication to customers. These techniques may include determining that a service health incident has customer impact, the service health incident corresponding to an outage of one or more services of a cloud computing platform, identifying a plurality of customers impacted by the service health incident, and predicting, based on the service health incident and one or more other service health incidents, aggregated incident information identifying a plurality of service health incidents associated with the outage of the one or more services. In addition, the techniques may include identifying the one or more services associated with the service health incident, and transmitting, based at least in part on the aggregated incident information and the one or more services, a health notification to the plurality of customers.