Abstract: A system for compressing data during neural network training, comprising of memory that stores computer executable components; a processor that executes computer executable components stored in the memory, wherein the computer executable components comprise of a compilation component that compiles respective molecular descriptors regarding a first set of molecules; a perception component that learns human perception information related to olfactory perceptions of the first set of molecules, and generates predictions of human olfactory perceptions of a second set of molecules; a fitting component that fits distance predictions from the perception component regarding the second set of molecules against measured correct classifications regarding the second set of molecules; and a vector component that generates a perceptual vector distance between two olfactory targets.

Abstract: A method, system and apparatus of using a computing device to explain one or more predictions of a machine learning model including receiving by a computing device a pre-trained artificial intelligence model with one or more predictions, generating by the computing device a multilevel explanation tree, linking neighborhood of datapoints around each of a plurality of training datapoints to the one or more predictions, and utilizing by the computing device the multilevel explanation tree to explain one or more predictions of the machine learning model.

Abstract: In an embodiment, a method for generating contrastive information for a classifier prediction comprises receiving image data representative of an input image, using a deep learning classifier model to predict a first classification for the input image, evaluating the input image using a plurality of classifier functions corresponding to respective high-level features to identify one or more of the high-level features absent from the input image, and identifying, from among the high-level features absent from the input image, a pertinent-negative feature that, if added to the input image, will result in the deep learning classifier model predicting a second classification for the modified input image, the second classification being different from the first classification. In an embodiment, the method includes creating a pertinent-positive image that is a modified version of the input image that has the first classification and fewer than all superpixels of the input image.

Abstract: Techniques facilitating interpretable rule generation using loss-preserving transformation are provided. In one example, a computer-implemented method can comprise evaluating, by a system operatively coupled to a processor, an input data set that comprises three data categories. The computer-implemented method can also comprise transforming, by the system, the input data set into a transformed data set. The transformed data set can comprise two data categories determined based on the three data categories. Transforming the input data set can comprise determining a first cost associated with the transformed data set is no greater than a second cost associated with the input data set.

Abstract: Techniques facilitating interpretable rule generation using loss-preserving transformation are provided. In one example, a computer-implemented method can comprise evaluating, by a system operatively coupled to a processor, an input data set that comprises three data categories. The computer-implemented method can also comprise transforming, by the system, the input data set into a transformed data set. The transformed data set can comprise two data categories determined based on the three data categories. Transforming the input data set can comprise determining a first cost associated with the transformed data set is no greater than a second cost associated with the input data set.

Abstract: A network is crawled using a trained learning model to identify a set of secondary-source documents related to an event. A hub page from the set of secondary-source documents is identified that includes a link predicted to link to a new relevant secondary-source document. The new document is added to the set of secondary-source documents. Information is extracted from the set of secondary-source documents. Feedback is received indicative of a relevancy level for the extracted information as applied to the event. Each document is classified into one or more categories related to the event, based on the extracted information and the received feedback information. A learning model is trained based on the received feedback.

Abstract: A method, system, and computer program product, including generating a contrastive explanation for a decision of a classifier trained on structured data, highlighting an important feature that justifies the decision, and determining a minimal set of new values for features that alter the decision.

Abstract: A method, system, and computer program product, including highlighting a minimally sufficient component in an input to justify a classification, identifying contrastive characteristics or features that are minimally and critically absent, maintaining the classification and distinguishing it from a second input that is closest to the classification but is identified as a second classification.

Abstract: A method, system, and computer program product, including generating, using a linear probe, confidence scores through flattened intermediate representations and theoretically-justified weighting of samples during a training of the simple model using the confidence scores of the intermediate representations.

Abstract: Predicting human olfactory perception based on molecular structure is described. Molecular descriptor data indicative of molecular descriptors associated with a group of molecular samples can be obtained. Olfactory perception indicator (OPI) data for a set of OPIs can also be obtained with respect to the molecular samples. A training model can be executed on the molecular descriptor data and the OPI data to yield an output model that correlates molecular attributes with OPIs for a single individual or across an aggregate of individuals. The output model can be used to predict olfactory perception for a particular compound or mixture based on which OPIs are correlated with molecular descriptors of the compound or mixture in the output model. The output model can also be inverted and used to identify molecular descriptors that are correlated with a desired set of OPIs. A molecular construct having the molecular descriptors can then be generated.

Abstract: A system for compressing data during neural network training, comprising of memory that stores computer executable components; a processor that executes computer executable components stored in the memory, wherein the computer executable components comprise of a compilation component that compiles respective molecular descriptors regarding a first set of molecules; a perception component that learns human perception information related to olfactory perceptions of the first set of molecules, and generates predictions of human olfactory perceptions of a second set of molecules; a fitting component that fits distance predictions from the perception component regarding the second set of molecules against measured correct classifications regarding the second set of molecules; and a vector component that generates a perceptual vector distance between two olfactory targets.

Abstract: An apparatus, method and computer program product for identifying fraud in transaction data. The method includes: receiving invoice data comprising a vendor, a requestor and events, receiving public data and private data sources, computing a vendor risk score using the public and private data sources matching the vendor of the invoice data, computing a requestor risk score using the public data sources and the private data sources matching the requestor of the invoice data, computing an active invoice score using the vendor risk score and the requestor risk score and when the active invoice score is greater than a predetermined amount, blocking the invoice data. In one embodiment, computing a vendor risk score comprises obtaining a weight and a confidence for the event, calculating an event vendor risk score using the weight times the confidence and combining the event vendor risk scores for all of the events.

Abstract: Techniques facilitating interpretable rule generation using loss-preserving transformation are provided. In one example, a computer-implemented method can comprise evaluating, by a system operatively coupled to a processor, an input data set that comprises three data categories. The computer-implemented method can also comprise transforming, by the system, the input data set into a transformed data set. The transformed data set can comprise two data categories determined based on the three data categories. Transforming the input data set can comprise determining a first cost associated with the transformed data set is no greater than a second cost associated with the input data set.

Abstract: Techniques facilitating interpretable rule generation using loss-preserving transformation are provided. In one example, a computer-implemented method can comprise evaluating, by a system operatively coupled to a processor, an input data set that comprises three data categories. The computer-implemented method can also comprise transforming, by the system, the input data set into a transformed data set. The transformed data set can comprise two data categories determined based on the three data categories. Transforming the input data set can comprise determining a first cost associated with the transformed data set is no greater than a second cost associated with the input data set.

Abstract: A network is crawled using a trained learning model to identify a set of secondary-source documents related to an event. A hub page from the set of secondary-source documents is identified that includes a link predicted to link to a new relevant secondary-source document. The new document is added to the set of secondary-source documents. Information is extracted from the set of secondary-source documents. Feedback is received indicative of a relevancy level for the extracted information as applied to the event. Each document is classified into one or more categories related to the event, based on the extracted information and the received feedback information. A learning model is trained based on the received feedback.

Abstract: Predicting human olfactory perception based on molecular structure is described. Molecular descriptor data indicative of molecular descriptors associated with a group of molecular samples can be obtained. Olfactory perception indicator (OPI) data for a set of OPIs can also be obtained with respect to the molecular samples. A training model can be executed on the molecular descriptor data and the OPI data to yield an output model that correlates molecular attributes with OPIs for a single individual or across an aggregate of individuals. The output model can be used to predict olfactory perception for a particular compound or mixture based on which OPIs are correlated with molecular descriptors of the compound or mixture in the output model. The output model can also be inverted and used to identify molecular descriptors that are correlated with a desired set of OPIs. A molecular construct having the molecular descriptors can then be generated.

Abstract: Systems, methods, and computer-readable media are described for predicting consumer response to a stimulus based on olfactory characteristics of the stimulus. An intrinsic factor score associated with a product can be determined based on an intrinsic attribute of the stimulus, and optionally, further based on data indicative of historical consumer response to olfactory characteristics of the stimulus. A social factor score associated with a user can also be determined using available olfactory preference data associated with the user and/or data representative of one or more social signals indicative of a predicted response of the user to olfactory characteristics of the stimulus. A collaborative filtering technique can be employed to determine a recommendation score for the stimulus using the intrinsic factor score and the social factor score. The recommendation score can be compared to a threshold value to determine whether to recommend the stimulus to the user.

Abstract: A method, a computer program product, and a computer system for identifying significant and consumable-insensitive trace features. A computer computes a residual in a first regression of one or more secondary factors on a target. The computer computes residuals in a second regression of the one or more secondary factors on each of one or more trace features in one or more trace feature sets. The computer computes, for the one or more trace feature sets, coefficients of determination in a third regression of the residuals in the second regression on the residual in the first regression. The computer ranks the one or more trace feature sets by sorting the coefficient of determination. The computer determines, based on rankings of the one or more trace feature sets, significant trace feature sets.

Abstract: An apparatus, method and computer program product for identifying fraud in transaction data. The method includes: receiving invoice data comprising a vendor, a requestor and events, receiving public data and private data sources, computing a vendor risk score using the public and private data sources matching the vendor of the invoice data, computing a requestor risk score using the public data sources and the private data sources matching the requestor of the invoice data, computing an active invoice score using the vendor risk score and the requestor risk score and when the active invoice score is greater than a predetermined amount, blocking the invoice data. In one embodiment, computing a vendor risk score comprises obtaining a weight and a confidence for the event, calculating an event vendor risk score using the weight times the confidence and combining the event vendor risk scores for all of the events.

Abstract: A method for detecting anomalous energy usage of building or household entities. The method applies a number of successively stringent anomaly detection techniques to isolate households that are highly suspect for having engaged in electricity theft via meter tampering. The system utilizes historical time series data of electricity usage, weather, and household characteristics (e.g., size, age, value) and provides a list of households that are worthy of a formal theft investigation. Generally, raw utility usage data, weather history data, and household characteristics are cleansed, and loaded into an analytics data mart. The data mart feeds four classes of anomaly detection algorithms developed, with each analytic producing a set of households suspected of having engaged in electricity theft. The system allows a user to select households from each list or a set based on the intersection of all individual sets.