Abstract: The present disclosure describes methods and systems, including computer-implemented methods, computer-program products, and computer systems, for automating a proactive Safety KPI analysis. Correlated data related to a safety key performance indicator (KPI) is obtained from a correlation engine. A safety KPI prediction related to safety incidents, future safety trends, or future safety KPIs is generated based on the received correlated data and at least one safety KPI prediction model. The generated safety KPI prediction is transmitted to a proactive monitoring and alerting engine and a safety KPI alert is generated based on the safety KPI prediction, at least one alert threshold, and the at least one safety KPI prediction model. Transmission of the generated safety KPI alert is then initiated.

Abstract: Systems and methods for implementing an artificially intelligent virtual assistant includes collecting a user query; using a competency classification machine learning model to generate a competency label for the user query; using a slot identification machine learning model to segment the text of the query and label each of the slots of the query; generating a slot value for each of the slots of the query; generating a handler for each of the slot values; and using the slot values to: identify an external data source relevant to the user query, fetch user data from the external data source, and apply one or more operations to the query to generate response data; and using the response data, to generate a response to the user query.

Abstract: Training method and apparatus for loan fraud detection model and a computer device are provided, wherein the training method for loan fraud detection model includes: acquiring identity information and user's bank statement information of a plurality of sample users, and fraud label information corresponding to each user; constructing an identity feature vector and a behavior pattern vector according to the identity information; constructing a statement feature vector according to the behavior pattern vector, a second vector transformation matrix and the user's bank statement information; generating a target feature vector according to the behavior pattern vector and the statement feature vector; feeding a target neural network with the target feature vector to acquire a fraud detection result of the target feature vector; and training the target neural network, the first vector transformation matrix and the second vector transformation matrix to obtain a loan fraud detection model.

Abstract: A computer generates labels for machine learning algorithms by retrieving, from a data storage circuit, multiple label sets that contain labels that each classify data points in a corpus of data. A graph is generated that includes a plurality of edges, each edge between two respective labels from different label sets of the multiple label sets. Weights are determined for the plurality of edges based upon a consistency between data points classified by two labels connected by the edges. An algorithm is applied that groups labels from the multiple label sets based upon the weights for the plurality of edges. Data points are identified from the corpus of data that represent conflicts within the grouped labels. An electronic message is transmitted in order to present the identified data points to entities for further classification. A new label set is generated using the further classification received from the entities.

Abstract: Embodiments of an adaptive virtual intelligent agent (“AVIA”) service are disclosed. It may include the functions of a human administrative assistant for an enterprise including customer support, customer relationship management, and fielding incoming caller inquiries. It also has multi-modal applications for the home through interaction with AVIA implemented in the home. It may engage in free-form natural language dialogs. During a dialog, embodiments maintain the context and meaning of the ongoing dialog and provides information and services as needed by the domain of the application. Over time, the service automatically extends its knowledge of the domain (as represented in the Knowledge Tree Graphs) through interaction with external resources.

Abstract: A system comprises a network of computers comprising a master computer and slave computers. For a machine learning problem that is partitioned into a number of correlated sub-problems, each master computer is configured to store tasks associated with the machine learning problem, and each of the slave computers is assigned one of the correlated sub-problems. Each slave computer is configured to store variables or parameters or both associated with the assigned one of the correlated sub-problems; obtain information about one or more tasks stored by the master computer without causing conflict with other slave computers with regard to the information; perform computations to update the obtained information and the variables or parameters or both of the assigned sub-problem; send the updated information to the master computer to update the information stored at the master computer; and store the updated variables or parameters or both of the assigned sub-problem.

Abstract: Provided is an information providing system for a system administrator not highly skilled in the arts of natural language or artificial intelligence to be able to easily correct or add a response to an input from a user to arrange the response into an appropriate one if the response is not appropriate. When an input sentence does not hit any knowledge data, the system administrator is able to add a combination of the input sentence and a response thereto by opening a new QA quick addition screen from a knowledge maintenance screen by pressing a link of the input sentence. Furthermore, the system administrator is able to easily add the input sentence to knowledge data as a synonymous sentence of a candidate Q, which is determined to be semantically close to the input sentence, by clicking the candidate Q on the knowledge maintenance screen.

Abstract: An information processing apparatus includes: a network information acquisition unit that acquires network information which includes target nodes and adjacent nodes; a classification ratio calculation unit that calculates a classification ratio, in which the target nodes are respectively classified as a plurality of communities corresponding to a predetermined number in the network information, so as to have correlation according to given resolutions with a classification ratio in which the adjacent nodes are respectively classified as the plurality of communities; a first type community generation unit that generates one or more first type communities; a classification ratio updating unit that updates the classification ratio relevant to the target nodes so as to have correlation with the classification ratio in which the adjacent nodes are respectively classified as the plurality of communities; and a second type community generation unit that generates one or more second type communities.

Abstract: A method of dynamically modifying target selection with a neural network includes dynamically modifying a selection function by controlling an amount of imbalance of connections in the neural network. A selected neuron represents one of multiple candidate targets.

Abstract: Various techniques are described for using machine-learning artificial intelligence to improve how trading data can be processed to detect improper trading behaviors such as trade spoofing. In an example embodiment, semi-supervised machine learning is applied to positively labeled and unlabeled training data to develop a classification model that distinguishes between trading behavior likely to qualify as trade spoofing and trading behavior not likely to qualify as trade spoofing. Also, clustering techniques can be employed to segment larger sets of training data and trading data into bursts of trading activities that are to be assessed for potential trade spoofing status.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a deep neural network. One of the methods includes generating a plurality of feature vectors that each model a different portion of an audio waveform, generating a first posterior probability vector for a first feature vector using a first neural network, determining whether one of the scores in the first posterior probability vector satisfies a first threshold value, generating a second posterior probability vector for each subsequent feature vector using a second neural network, wherein the second neural network is trained to identify the same key words and key phrases and includes more inner layer nodes than the first neural network, and determining whether one of the scores in the second posterior probability vector satisfies a second threshold value.

Abstract: Providing a view of relevant items of a content collection includes identifying a current context based temporal parameters, spatial parameters, navigational parameters, lexical parameters, organizational parameters, and/or events, evaluating each of the items of the content collection according to the current context to provide a value for each of the items, and displaying a subset of the items corresponding to highest determined values. The temporal parameters may include a time of recent access of an item, frequency of access of an item, frequency of location related access of an item, and frequency of event related access of an item. Temporal patterns of accessing items may be numerically assessed based on time of day, time of week, and/or time of month. Evaluating each item may include determining a distance from a separating hyperplane using a support vector machine classification method.

Abstract: Relationship building Websites collect considerable self-reported and autonomously collected attribute data on users. Attribute data may be useful for identifying users having compatible or potentially compatible interests, likes, goals, and/or aspirations that the formation of a relationship between the users is possible. At least a portion of the data collected by relationship building Websites may include inbound and outbound messaging statistics and behaviors. When used in conjunction with profile attributes, these messaging statistics and behaviors may be used as training data to generate one or more response predictive models that provide an indication of the profile attributes and messaging behaviors to which a particular user is most likely to respond. Since messaging traffic is a key indicator of relationship building Website health and vitality, it is advantageous to provide users with matches or potential matches with whom they are more likely to exchange messages.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for parallel time series analysis. One of the methods includes receiving a plurality of data elements. The plurality of data elements are partitioned into a plurality of chunks, wherein the plurality of chunks, including a first chunk and a plurality of additional chunks, have an ordering according to the data elements included in each chunk. Each chunk is assigned to a particular segment of a plurality of segments. A first iteration of an autoregressive integrated moving average is computed for each chunk assigned to each segment. A second iteration of the autoregressive integrated moving average is computed for each chunk assigned to each segment, wherein computing uses the result data for a corresponding preceding chunk in the first iteration. One or more additional iterations of the autoregressive integrated moving average are computed until stopping criteria has been satisfied.

Abstract: Systems and methods are provided for accurately setting notification priority levels for applications on a user device. An example system includes a user device and a management server. When an application generates a notification, it provides a priority level for the notification. A management agent executing on the user device can detect the notification and its assigned priority level, determine a predicted priority level using a prediction engine or prediction server, and cause the application the replace or update the assigned priority level based on the predicted priority level. The management agent can then receive user actions related to that notification from the application, and use that information to determine an observed priority level. The prediction engine or prediction server can be updated based on the observed priority level, thereby dynamically increasing the accuracy of predictions.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for adversarial training of a neural network. One of the methods includes obtaining a plurality of training inputs; and training the neural network on each of the training inputs, comprising, for each of the training inputs: processing the training input using the neural network to determine a neural network output for the training input; applying a perturbation to the training input to generate an adversarial perturbation of the training input; processing the adversarial perturbation of the training input using the neural network to determine a neural network output for the adversarial perturbation; and adjusting the current values of the parameters of the neural network by performing an iteration of a neural network training procedure to optimize an adversarial objective function.

Abstract: A computer-implemented method for representation of weight values in an artificial neural network using inter-group indexing may include, in an artificial neural network that includes neurons and synaptic connections between the neurons with each of the synaptic connections including a weight, arranging the weights in ascending order. The method may also include dividing the arranged weights into groups based on approximately linear patterns of the weights. The method may further include designating a base group and the other groups as dependent groups. The method may also include storing in memory values of the weights in the base group. The method may further include storing in the memory a group index for each of the dependent groups. The method may also include storing in the memory an index for each of the weights in the dependent groups corresponding to one of the weights in the base group without storing in the memory values of the weights in the dependent groups.

Abstract: A method, system and computer program product for generating a solution to an optimization problem. A received structured set of data is analyzed with the prescriptive domains to identify one or more prescriptive domains that match the received structure set of data in data structure and/or semantic terms. A user selection of one of the presented possible prescriptive intentions from the intention templates in the identified one or more prescriptive domains that match the received structure set of data in data structure and/or semantic terms is received. A prescriptive model is then generated from the prescriptive domain containing the selected prescriptive intention. The prescriptive model is translated into a technical prescriptive model using a set of mapping rules. Furthermore, the technical prescriptive model is translated into an optimization model. The optimization model is solved and an output defining a solution from the solved optimization model is presented.

Abstract: Techniques in advanced deep learning provide improvements in one or more of accuracy, performance, and energy efficiency. An array of processing elements performs flow-based computations on wavelets of data. Each processing element has a compute element with dedicated storage and a routing element. Each router enables communication with nearest neighbors in a 2D mesh. The communication is via wavelets in accordance with a representation comprising an index specifier, a virtual channel specifier, a task specifier, a data element specifier, and an optional control/data specifier. The virtual channel specifier and the task specifier are associated with one or more instructions. The index specifier and the data element are optionally associated with operands of the one or more instructions.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a plurality of different types of predictive models using training data, wherein each of the predictive models implements a different machine learning technique. One or more weights are obtained wherein each weight is associated with an answer category in the plurality of examples. A weighted accuracy is calculated for each of the predictive models using the one or more weights.