Abstract: Techniques for generating machine learning training data which corresponds to one or more downstream tasks are disclosed. In one example, a computer implemented method comprises generating one or more synthetic data instances for training a machine learning model, and determining a value of respective ones of the one or more synthetic data instances with respect to at least one task. One or more additional synthetic data instances for training the machine learning model are generated based at least in part on the values of the respective ones of the one or more synthetic data instances.

Abstract: Methods, systems, and computer program products for dynamic pricing of energy consumed from a shared battery using real-time consumption data are provided herein. A computer-implemented method includes calculating wear cost arising from a battery shared by multiple users, wherein the wear cost is based on usage data of the battery; calculating a proportionality factor for each of the multiple users for the calculated wear cost, wherein the proportionality factor is based on individual usage of the battery; apportioning the calculated wear cost to each of the multiple users based on the user's proportionality factor; and determining a dynamic price for energy used by the battery for each of the multiple users, based on said apportioning.

Abstract: Methods, systems, and computer program products for constraints-based refactoring of monolith applications through attributed graph embeddings are provided herein.

Abstract: One embodiment provides a method, including: receiving information describing an application to be split into a plurality of microservices; identifying, utilizing a microservices advisor application, application elements of the application; generating, utilizing the microservices advisor application and from the application elements, a heterogenous graph, wherein each node within the heterogenous graph represents an application element and wherein each edge within the heterogenous graph represents a relationship between two nodes connected by the edge; identifying, based upon user input identifying preferences of relationships between nodes, groups of nodes within the heterogenous graph sharing a common attribute; and providing, from the microservices advisor application, a recommendation, based upon the identified groups of nodes, for splitting the application into microservices, wherein the recommendation includes a number of microservices for the application and application elements that should be included

Abstract: Methods, systems, and computer program products for automatically detecting periods of normal activity by analyzing observability data in IT operations environments are provided herein. A computer-implemented method includes obtaining multiple types of data related to one or more artificial intelligence-related information technology operations; modelling at least a portion of the obtained data as time series data; automatically identifying, from the time series data, one or more time periods associated with one or more given levels of data activity; and performing one or more automated actions, in at least one artificial intelligence-related information technology operations environment, based at least in part on the data corresponding to the one or more identified time periods.

Abstract: Methods, systems, and computer program products for federated machine learning based on partially secured spatio-temporal data are provided herein. A computer-implemented method includes obtaining temporal data from a plurality of distributed client devices in conjunction with a federated machine learning process, wherein at least a portion of the data comprises encoded private data and at least a portion of the data is public data; generating a spatio-temporal graph comprising nodes representing the plurality of distributed client devices, wherein the generating comprises identifying at least one pair of similar nodes based at least in part on the public data and adding an edge to the spatio-temporal graph between the pair of similar nodes; and aligning encoders of at least two of the distributed client devices based on the spatio-temporal graph.

Abstract: Techniques for generating machine learning training data which corresponds to one or more downstream tasks are disclosed. In one example, a computer implemented method comprises generating one or more synthetic data instances for training a machine learning model, and determining a value of respective ones of the one or more synthetic data instances with respect to at least one task. One or more additional synthetic data instances for training the machine learning model are generated based at least in part on the values of the respective ones of the one or more synthetic data instances.

Abstract: One embodiment provides a method, including: receiving information describing an application to be split into a plurality of microservices; identifying, utilizing a microservices advisor application, application elements of the application; generating, utilizing the microservices advisor application and from the application elements, a heterogenous graph, wherein each node within the heterogenous graph represents an application element and wherein each edge within the heterogenous graph represents a relationship between two nodes connected by the edge; identifying, based upon user input identifying preferences of relationships between nodes, groups of nodes within the heterogenous graph sharing a common attribute; and providing, from the microservices advisor application, a recommendation, based upon the identified groups of nodes, for splitting the application into microservices, wherein the recommendation includes a number of microservices for the application and application elements that should be included

Abstract: Methods, systems, and computer program products for constraints-based refactoring of monolith applications through attributed graph embeddings are provided herein.

Abstract: A dynamic graph includes a plurality of nodes and edges at a plurality of time steps; each node corresponds to a geographic location in a first area where pest infestation information is available for a subset of locations. Each edge connects two of the nodes which are geographically proximate, has a direction based on wind direction, and has a weight based on relative wind speed. Assign node features based on weather data as well as labels corresponding to pest infestation severity. Train a graph convolutional network on the dynamic graph. Based on predicted future weather conditions for a second area different than the first area, use the trained graph convolutional network to predict, via inductive learning, pest infestation severity for future times for a new set of nodes corresponding to new geographic locations in the second area for which no pest infestation information is available.

Abstract: Methods, systems, and computer program products for automatically detecting outliers in federated data are provided herein. A computer-implemented method includes obtaining local outlier-related data from multiple client systems within a federated learning environment; detecting one or more federated learning environment-level outliers from at least a portion of the multiple client systems by processing at least a portion of the obtained local outlier-related data using one or more artificial intelligence models; determining at least one calibration parameter for detecting federated learning environment-level outliers based at least in part on the one or more detected federated learning environment-level outliers; and outputting the at least one determined calibration parameter to at least a portion of the multiple client systems within the federated learning environment.

Abstract: Methods, systems, and computer program products for dynamic pricing of energy consumed from a shared battery using real-time consumption data are provided herein. A computer-implemented method includes calculating wear cost arising from a battery shared by multiple users, wherein the wear cost is based on usage data of the battery; calculating a proportionality factor for each of the multiple users for the calculated wear cost, wherein the proportionality factor is based on individual usage of the battery; apportioning the calculated wear cost to each of the multiple users based on the user's proportionality factor; and determining a dynamic price for energy used by the battery for each of the multiple users, based on said apportioning.

Abstract: One embodiment provides a method, including: training a plurality of machine-learning models, wherein each of the machine-learning models is trained for a specific farm field region utilizing training data for the plurality of machine-learning models; wherein the utilizing training data includes identifying one of the farm field regions having a similarity to another of the farm field regions and transferring training data; identifying a plurality of types of data needed for updating at least one of the plurality of machine-learning models to address at least one uncertainty; recommending collection of and collecting at least one of the plurality of types of data; and re-training the subset of the plurality of machine-learning models utilizing the at least one of the plurality of types of data, thereby decreasing the cost the of data collection, for example, crowdsourced data, by utilizing data collected from one farm field region in other farm field regions.

Abstract: A dynamic graph includes a plurality of nodes and edges at a plurality of time steps; each node corresponds to a geographic location in a first area where pest infestation information is available for a subset of locations. Each edge connects two of the nodes which are geographically proximate, has a direction based on wind direction, and has a weight based on relative wind speed. Assign node features based on weather data as well as labels corresponding to pest infestation severity. Train a graph convolutional network on the dynamic graph. Based on predicted future weather conditions for a second area different than the first area, use the trained graph convolutional network to predict, via inductive learning, pest infestation severity for future times for a new set of nodes corresponding to new geographic locations in the second area for which no pest infestation information is available.

Abstract: One embodiment provides a method, including: identifying a target farmer having a farm producing a crop; receiving at least one target farmer input related to at least one of: (i) a growing condition of the crop of the target farmer and (ii) a characteristic of the crop; receiving, from each of the plurality of other farmers, additional input; validating the at least one target farmer input by comparing the at least one target farmer input to the additional inputs; generating a reliability score for the target farmer; and producing a ranking for the target farmer with respect to the plurality of other farmers.

Abstract: Methods, systems, and computer program products for prioritizing errors in connectivity models of distribution networks are provided herein.

Abstract: Methods, systems, and computer program products for mapping wind turbines and predicting wake effects using satellite imagery data are provided herein. A computer-implemented method includes analyzing one or more satellite images depicting one or more portions of a pre-determined geographic area; detecting a group of one or more wind turbines in the pre-determined geographic area based on the analyzing step and one or more additional items of data; inferring geographic coordinates of each of the detected wind turbines; predicting a wake effect impacting one or more of the detected wind turbines based on the inferred geographic coordinates of each of the detected wind turbines and forecasted weather data; and outputting the predicted wake effect to at least one user.

Abstract: Methods, systems, and computer program products for detecting losses in electrical networks are provided herein. A computer-implemented method includes computing a consumption estimation for each consumer associated with a network; determining a difference between (i) the consumption estimation and (ii) actual consumption for each consumer; clustering the consumers into a cluster based on a consumption pattern associated with each consumer; determining a level of deviation of (i) the consumption pattern associated with each consumer from (ii) a consumption pattern representative of the cluster; clustering the consumers into two or more clusters based on a consumption pattern during a first interval of time and during a second interval of time; determining, for each consumer, a level of evolution from (i) a first cluster during the first interval to (ii) a second cluster during the second interval; and identifying consumers associated with a given loss within the network.

Abstract: A method of two-factor authentication for gaining access to an application using at least a first device and a second device registered to a user. The first and second devices each have a plurality of sensors for detecting activity modalities of the user on the first and the second devices and are in communication with a server computer. In the method, the server computer: receives credentials and detected activity modality for gaining access to the application from the first device; sends a request to the second device registered to the user for activity modality of the user; receives the detected activity modality from the second device; compares the detected activity modality of first device to the detected activity modality of the second device; and if the detected activity modalities of first device and the second device match, granting access to the user on the first device to the application.

Abstract: Methods, systems, and computer program products for mapping wind turbines and predicting wake effects using satellite imagery data are provided herein. A computer-implemented method includes analyzing one or more satellite images depicting one or more portions of a pre-determined geographic area; detecting a group of one or more wind turbines in the pre-determined geographic area based on the analyzing step and one or more additional items of data; inferring geographic coordinates of each of the detected wind turbines; predicting a wake effect impacting one or more of the detected wind turbines based on the inferred geographic coordinates of each of the detected wind turbines and forecasted weather data; and outputting the predicted wake effect to at least one user.