Abstract: Methods, systems, and computer program products for providing the status of model extraction in the presence of colluding users are provided herein. A computer-implemented method includes generating, for each of multiple users, a summary of user input to a machine learning model; comparing the generated summaries to boundaries of multiple feature classes within an input space of the machine learning model; computing correspondence metrics based at least in part on the comparisons; identifying, based at least in part on the computed metrics, one or more of the multiple users as candidates for extracting portions of the machine learning model in an adversarial manner; and generating and outputting an alert, based on the identified users, to an entity related to the machine learning model.

Abstract: Provided are techniques for global context explainers for Artificial Intelligence systems using multivariate timeseries data. Predictions for multivariate timeseries data are received. Feature importance weights are generated from the predictions using a feature-based local explainer, where each of the feature importance weights is associated with a time period and a corresponding data source of timeseries data of the multivariate timeseries data. A dataset is generated using the feature importance weights, where the dataset includes, for each time period and the corresponding data source, a label indicating whether the feature importance weight is one of positive and negative. One or more global explanations are generated using the dataset and a directly interpretable rule-based explainer, where the one or more global explanations indicate how the predictions change at particular times in the multivariate timeseries data based on values from the corresponding data source.

Abstract: In an embodiment, a method for inspecting and transforming a machine learning model includes receiving a request that includes the machine learning model and a configuration object that provides an indication of a selected strategy. In the embodiment, the method includes creating a partially specified task graph that includes a first placeholder node for a future expanded task node. In the embodiment, the method includes performing a dynamic expansion and execution phase that includes, repeatedly (a) using a cognitive engine to evaluate whether to revise the partially specified task graph based at least in part on the selected strategy, and (b) using a processor-based execution engine to perform an action specified by the complete node. In an embodiment, the dynamic expansion and execution phase repeats until after the cognitive engine adds a consolidated results node.

Abstract: Methods, systems, and computer program products for generating a framework for prioritizing machine learning model offerings via a platform are provided herein. A computer-implemented method includes processing, via a computing platform, a machine learning model input by a first user and metadata corresponding to the machine learning model input by the first user; automatically comparing, via the computing platform, the metadata corresponding to the machine learning model with metadata corresponding to one or more existing machine learning models stored by the computing platform; automatically calculating, via the computing platform, initial pricing information for the machine learning model based on the comparison; and outputting, via an interactive user interface of the computing platform, the machine learning model to one or more additional users for purchase in accordance with the calculated initial pricing information.

Abstract: One embodiment provides a method for recommending model characteristics to be used in developing a target geo-spatial physical model for a target geographic location utilizing historical lineage data corresponding to historical geo-spatial physical models, including: receiving information related to the target geographic location, wherein the information describes geographical and domain features of the target geographic location; identifying, using at least one similarity algorithm, at least one other geographic location that is similar to the target geographic location, wherein the at least one geographic location has at least one corresponding historical geo-spatial physical model; and recommending, using at least one machine-learning model and based upon the at least one other geographic location, initial model characteristics for developing and deploying the target geo-spatial physical model.

Abstract: One embodiment provides a method, including: accessing historical deployment information for a plurality of deep-learning models, wherein the historical deployment information identifies values for model parameters of a deep-learning model during deployment of the deep-learning model; receiving information related to a target deep-learning model that a developer is creating, wherein the received information identifies components being utilized in the target deep-learning model; determining, by comparing the received information to the historical deployment information, expected values for target model parameters of the target deep-learning model based upon the components utilized within the target deep-learning model; and providing a recommendation for a modification to the target deep-learning model based upon the expected values, wherein the modification comprises a change to at least one component of the target deep-learning model.

Abstract: An example operation may include one or more of receiving, at a node, a request to execute a software model that has been decomposed into a plurality of sequential sub-components, executing a sub-component from among the plurality of sub-components based on input data included in the received request to generate output data, hashing the input data and the output data to generate a hashed execution result of the sub-component, and storing the hashed execution result of the sub-component within a block among a hash-linked chain of blocks which include hashed execution results of other sub-components of the software model executed by other nodes.

Abstract: In an embodiment, a method for inspecting and transforming a machine learning model includes receiving a request that includes the machine learning model and a configuration object that provides an indication of a selected strategy. In the embodiment, the method includes creating a partially specified task graph that includes a first placeholder node for a future expanded task node. In the embodiment, the method includes performing a dynamic expansion and execution phase that includes, repeatedly (a) using a cognitive engine to evaluate whether to revise the partially specified task graph based at least in part on the selected strategy, and (b) using a processor-based execution engine to perform an action specified by the complete node. In an embodiment, the dynamic expansion and execution phase repeats until after the cognitive engine adds a consolidated results node.

Abstract: Methods, systems, and computer program products for generating a framework for prioritizing machine learning model offerings via a platform are provided herein. A computer-implemented method includes processing, via a computing platform, a machine learning model input by a first user and metadata corresponding to the machine learning model input by the first user; automatically comparing, via the computing platform, the metadata corresponding to the machine learning model with metadata corresponding to one or more existing machine learning models stored by the computing platform; automatically calculating, via the computing platform, initial pricing information for the machine learning model based on the comparison; and outputting, via an interactive user interface of the computing platform, the machine learning model to one or more additional users for purchase in accordance with the calculated initial pricing information.

Abstract: One embodiment provides a method for delaying malicious attacks on machine learning models that a trained using input captured from a plurality of users, including: deploying a model, said model designed to be used with an application, for responding to requests received from users, wherein the model comprises a machine learning model that has been previously trained using a data set; receiving input from one or more users; determining, using a malicious input detection technique, if the received input comprises malicious input; if the received input comprises malicious input, removing the malicious input from the input to be used to retrain the model; retraining the model using received input that is determined to not be malicious input; and providing, using the retrained model, a response to a received user query, the retrained model delaying the effect of malicious input on provided responses by removing malicious input from retraining input.

Abstract: Methods, systems, and computer program products for cognitively predicting dust deposition on solar photovoltaic modules are provided herein. A computer-implemented method includes deriving, with respect to solar photovoltaic modules, dust parameters from image data, and estimating, for a given future time at a current module orientation, an amount of surface area of the modules that will be covered by dust and a yield loss of the modules associated with dust coverage. The method also includes forecasting, for the given future time at each of one or more modified module orientations, an amount of surface area of the modules that will be covered by dust and a yield loss of the modules associated with dust coverage. Further, the method includes generating an instruction to change the orientation of at least one of the modules, and outputting the instruction to at least one actuation system associated with the modules.

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

Abstract: According to one or more embodiments, a method, a computer program product, and a computer system for managing vegetation across distribution systems are provided. The method may include receiving, by a computer, voltage data from one or more data sensors. The computer may determine one or more locations of one or more voltage fault conditions based on the received voltage data. A score may be assigned to each of the determined locations by the computer. The computer may then identify a subset of one or more work orders corresponding to the one or more determined locations from among a database of work orders. A subset of locations may be determined by the computer from among the one or more locations based on the assigned scores and the identified subset of work orders. A field visit may then be scheduled by the computer based on the determined subset of locations.

Abstract: An example operation may include one or more of receiving, at a node, a request to execute a software model that has been decomposed into a plurality of sequential sub-components, executing a sub-component from among the plurality of sub-components based on input data included in the received request to generate output data, hashing the input data and the output data to generate a hashed execution result of the sub-component, and storing the hashed execution result of the sub-component within a block among a hash-linked chain of blocks which include hashed execution results of other sub-components of the software model executed by other nodes.

Abstract: One embodiment provides a method for delaying malicious attacks on machine learning models that a trained using input captured from a plurality of users, including: deploying a model, said model designed to be used with an application, for responding to requests received from users, wherein the model comprises a machine learning model that has been previously trained using a data set; receiving input from one or more users; determining, using a malicious input detection technique, if the received input comprises malicious input; if the received input comprises malicious input, removing the malicious input from the input to be used to retrain the model; retraining the model using received input that is determined to not be malicious input; and providing, using the retrained model, a response to a received user query, the retrained model delaying the effect of malicious input on provided responses by removing malicious input from retraining input.

Abstract: Methods, systems, and computer program products for providing the status of model extraction in the presence of colluding users are provided herein. A computer-implemented method includes generating, for each of multiple users, a summary of user input to a machine learning model; comparing the generated summaries to boundaries of multiple feature classes within an input space of the machine learning model; computing correspondence metrics based at least in part on the comparisons; identifying, based at least in part on the computed metrics, one or more of the multiple users as candidates for extracting portions of the machine learning model in an adversarial manner; and generating and outputting an alert, based on the identified users, to an entity related to the machine learning model.

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: One embodiment provides a method including: prior to an initial period of operation of an appliance, storing in memory a first set of characteristics of the appliance; during an initial period of operation of the appliance, learning a second set of characteristics of the appliance; during subsequent operation of the appliance: detecting an adverse operating condition of the appliance; and based on the first set of characteristics, the second set of characteristics and the detected adverse operating condition, determining a corrective action to be taken with regard to the appliance, the corrective action comprising at least one of: switching off the appliance and warning a user of the detected adverse operating condition. Other aspects are described and claimed.

Abstract: Methods, systems, and computer program products for cognitively predicting dust deposition on solar photovoltaic modules are provided herein. A computer-implemented method includes deriving, with respect to solar photovoltaic modules, dust parameters from image data, and estimating, for a given future time at a current module orientation, an amount of surface area of the modules that will be covered by dust and a yield loss of the modules associated with dust coverage. The method also includes forecasting, for the given future time at each of one or more modified module orientations, an amount of surface area of the modules that will be covered by dust and a yield loss of the modules associated with dust coverage. Further, the method includes generating an instruction to change the orientation of at least one of the modules, and outputting the instruction to at least one actuation system associated with the modules.

Abstract: According to one or more embodiments, a method, a computer program product, and a computer system for managing vegetation across distribution systems are provided. The method may include receiving, by a computer, voltage data from one or more data sensors. The computer may determine one or more locations of one or more voltage fault conditions based on the received voltage data. A score may be assigned to each of the determined locations by the computer. The computer may then identify a subset of one or more work orders corresponding to the one or more determined locations from among a database of work orders. A subset of locations may be determined by the computer from among the one or more locations based on the assigned scores and the identified subset of work orders. A field visit may then be scheduled by the computer based on the determined subset of locations.