Abstract: This disclosure relates to method and system for improving performance of an artificial neural network (ANN). The method may include generating a weight matrix comprising weights of neural nodes in a given layer for each layer of the ANN, determining a marginal contribution value of a given neural node for each neural node in the given layer with respect to other neural nodes in the given layer, executing an elimination decision for each neural node in each layer based on the corresponding marginal contribution value, determining a distributed surplus value of a given remaining neural node in a given layer based on the marginal contribution values of a coalition of remaining neural nodes in the given layer for each remaining neural node in each layer, and updating the weight matrix based on the distributed surplus value of each remaining neural node in each layer.

Abstract: Provided is a process, including: obtaining a first training dataset of subject-entity records; training a first machine-learning model on the first training dataset; forming virtual subject-entity records by appending members of a set of candidate action sequences to time-series of at least some of the subject-entity records; forming a second training dataset by labeling the virtual subject-entity records with predictions of the first machine-learning model; and training a second machine-learning model on the second training dataset.

Abstract: Techniques and systems for critical dimension metrology are disclosed. Critical parameters can be constrained with at least one floating parameter and one or more weight coefficients. A neural network is trained to use a model that includes a Jacobian matrix. During training, at least one of the weight coefficients is adjusted, a regression is performed on reference spectra, and a root-mean-square error between the critical parameters and the reference spectra is determined. The training may be repeated until the root-mean-square error is less than a convergence threshold.

Abstract: Systems and methods for modelling time-series data includes testing a testing model with a plurality of hyper-forgetting rates to select a best performing hyper-forgetting rate. A model optimization is tested using the best performing hyper-forgetting rate with the testing model to test combinations of hyper-parameters to select a best performing combination of hyper-parameters. An error of the model is determined using the model optimization. Model parameters are recursively updated according to the least squares regression by determining a pseudo-inverse of a Hessian of the least squares regression at a current time stamp according to a projection of the time-series data at the current time stamp and the pseudo-inverse of the Hessian at a previous time-stamp to determine an optimum model parameter. A next step behavior of the time-series data is predicted using the optimum model parameter. The next step behavior is stored in a database for access by a user.

Abstract: Disclosed is a system and a method for collaborative decision making in dynamically changing environment. A query corresponding to a problem is received from a user. Further, one or more intermediate steps required to reach a decision is calculated based on metadata associated to the problem. A decision-making flow is established for the one or more intermediate steps required to reach the decision. It may be noted that the decision-making flow corresponds to a sequence for execution of the one or more intermediate steps. Further, a decision space comprising one or more decision options is generated. The decision space is dynamically modified based on one or more uncertain events. A decision knowledge graph depicting modifications in the decision space is generated. Further, the decision space and the decision knowledge graph are updated. Finally, the decision is selected based on the updated decision knowledge graph and the updated decision space.

Abstract: Embodiments of the present invention disclose a method, a computer program product, and a computer system predicting parking availability. A computer identifies parking spaces and groups the parking spacing into parking locations. In addition, the computer distinguishes private parking spaces from public parking spaces, and trains a crowd forecast model for each of the parking locations. The computer further receives a destination and preferences, from which the computer creates a geofence based on the destination and preferences. The computer then predicts parking availability based on the crowd forecast models and refines the crowd forecast model.

Abstract: A system and method for generating predictions of geopolitical events is provided. Predictions may be generated by retrieving relevant metadata associated with a content item and evaluating one or more signals representative of the same.

Abstract: The disclosed embodiments concern methods, apparatus, systems and computer program products for determining and displaying pedigrees based on IBD data. Some implementations use a probabilistic relationship model to obtain various likelihoods of various potential relationships based on pairwise IBD data and pairwise age data. Some implementations build large pedigrees by combining smaller pedigrees. Some implementations display pedigree graphs with various features that are informative and easy to understand.

Abstract: An artificial intelligence device is disclosed. In an embodiment, the artificial intelligence device includes a sensor configured to acquire an output value according to control of a control system, and an artificial intelligence unit comprising one or more processors configured to obtain one or more updated parameters of a control function of the control system based on the output value using reinforcement learning, and update the control function for providing a control value to the control system with the one or more updated parameters.

Abstract: A computer processor generates a topic-based dataset based on parsing content received from a plurality of information sources, which includes historical data and scientific data, associated with a location of a natural resource. The processor generates a plurality of clusters, respectively corresponding to like-topic data of the topic-based dataset. The processor determines a plurality of hypotheses, respectively corresponding to the plurality of clusters of the like-topic data, wherein the plurality of hypotheses are based on features associated with each of the plurality of clusters of the like-topic data. The processor combines pairs of clusters, based on a similarity heuristic applied to the one or more pairs of clusters, and the processor determines a plurality of probabilities respectively corresponding to a validity of each hypothesis of the plurality of hypotheses, associated with the location of a natural resource.

Abstract: An information processing apparatus includes a history acquisition section configured to acquire history data including a history indicating that a plurality of selection subjects have selected selection objects; a learning processing section configured to allow a choice model to learn a preference of each selection subject for a feature and an environmental dependence of selection of each selection object in each selection environment using the history data, where the choice model uses a feature value possessed by each selection object, the preference of each selection subject for the feature, and the environmental dependence indicative of ease of selection of each selection object in each of a plurality of selection environments to calculate a selectability with which each of the plurality of selection subjects selects each selection object; and an output section configured to output results of learning by the learning processing section.

Abstract: Querying databases may be performed with references to machine learning models. A database query may be received that references a machine learning model and database. In response to the query, the machine learning model may provide information which may be returned as part of a result of the query or may be used to generate a result of the query. The machine learning model may be generated in response to a request to generate a machine learning model that includes a database query that identifies the data upon which a machine learning technique may be applied to generate the machine learning model.

Abstract: Devices and methods for performing computations of a neural network include determining, for a particular layer of the neural network, a number of input channels and a number of output maps generated for particular pixels of a plurality of pixels. A portion of a processing chip is configured into a plurality of processing units, with the processing chip including a plurality of tensor arrays and a plurality memory cells. A particular processing unit of the plurality of processing units performs computations associated with a particular pixel. Individual processing units of the plurality of processing units each include a number of tensor arrays determined based on the number of input channels and a number of memory cells or pixel arrays corresponding to a number of output maps. The plurality of processing units is assigned to perform computations of the particular layer.

Abstract: Examples include techniques to manage training or trained models for deep learning applications. Examples include routing commands to configure a training model to be implemented by a training module or configure a trained model to be implemented by an inference module. The commands routed via out-of-band (OOB) link while training data for the training models or input data for the trained models are routed via inband links.

Abstract: An apparatus and method are provided for machine learning method using a network of agents. The agents are arranged in a network with respective links between pairs of agents, and the links enabling the exchange information. Different agents can apply different reasoning paradigms corresponding to different approaches to machine learning and artificial intelligence. These disparate approaches are seamlessly integrated to aggregate decisions and learning performed using different approaches using an economics model in which a Nash equilibrium is reached through the exchange of information. Each agent selects which other agents to exchange information with by seeking to optimize preference, utility, and objective functions, and these function include how well the agents obtain an assigned goal subject to other desirable features and characteristics (e.g., enforcing diversity).

Abstract: A method, performed by an electronic device, of predicting an electronic structure of a material includes: receiving input data of a user related to elements constituting the first material; applying the received input data to a trained model for estimating a density of state of the first material; and outputting a first graph indicating the density of state for each energy level of the first material output from the trained model, wherein the trained model is trained to generate the first graph based on pre-input data about a plurality of second materials composed of at least some of the elements constituting the first material and a plurality of second graphs representing the density of state for each energy level of the plurality of second materials.

Abstract: Dialog system training techniques using a simulated user system are described. In one example, a simulated user system supports multiple agents. The dialog system, for instance, may be configured for use with an application (e.g., digital image editing application). The simulated user system may therefore simulate user actions involving both the application and the dialog system which may be used to train the dialog system. Additionally, the simulated user system is not limited to simulation of user interactions by a single input mode (e.g., natural language inputs), but also supports multimodal inputs.

Abstract: In a system for knowledge system management, the system includes: a processor; and a memory coupled to the processor, wherein the memory has stored thereon instructions that, when executed by the processor, cause the processor to: receive a question from a customer; determine whether to not the question satisfies a first criterion for obtaining an automated response to the question; in response to determining that the question satisfies the first criterion, transmit the question to a knowledge system for generating automated responses to questions; receive the automated response to the question from the knowledge system; determine whether or not the automated response satisfies a second criterion for providing the automated response to the customer; and in response to determining that the automated response satisfies the second criterion, transmit the automated response to an electronic device operated by the customer.

Abstract: A model visualizer visualizes a neural network model at a neuron level. The model visualizer receives a plurality of instances comprising a plurality of features, and receives a neural network model comprising a plurality of layers, each layer comprising a plurality of neurons. For each neuron of each layer, the model visualizer selects a group of instances that highly activate that neuron to generate a summary feature vector, calculates an average feature value of the summary feature vector, compares each feature value of the summary feature vector with the average feature value, and assigns an attribute to each element of the summary feature vector based on the comparison. The model visualizer visualizes the received neural network model based on the assignment to each neuron.

Abstract: A machine learning (“ML”) pipeline that includes unsupervised learning, supervised learning, and Bayesian learning is utilized to train a ML classifier that can classify machine metrics as being indicative of an anomaly. A boosting process can be utilized during the unsupervised learning portion of the ML pipeline that scores clusters of training data for completeness, and further splits clusters of training data based upon the completeness scores in order to optimize the clustering of the training data. Supervised learning is then performed on the cluster-labeled training data. Bayesian learning can also be utilized to assign incident probability inferences to the clusters of training data. Once the ML classifier has been trained, the ML classifier can be utilized in a production environment to classify multi-dimensional machine metrics generated by computing devices in the production environment as being indicative of an anomaly.