Abstract: Techniques to correlate event data are disclosed. In various embodiments, an aggregation engine is used to correlate event data received from one or more source systems based on one or more correlation rules. An event group comprising at least a portion of said correlated event data is generated programmatically and is asserted as a fact in a working memory of a Rete engine configured to apply one or more Rete rules to facts in the working memory.

Abstract: The present disclosure provides systems and methods to reduce computational costs associated with convolutional neural networks. In addition, the present disclosure provides a class of efficient models termed “MobileNets” for mobile and embedded vision applications. MobileNets are based on a straight-forward architecture that uses depthwise separable convolutions to build light weight deep neural networks. The present disclosure further provides two global hyper-parameters that efficiently trade-off between latency and accuracy. These hyper-parameters allow the entity building the model to select the appropriately sized model for the particular application based on the constraints of the problem. MobileNets and associated computational cost reduction techniques are effective across a wide range of applications and use cases.

Abstract: Systems and methods are provided to perform weight update operations in a resistive processing unit (RPU) system to update weight values of RPU devices comprising tunable resistive device. A weight update operation for a given RPU device includes maintaining a weight update accumulation value for the RPU device, adjusting the weight update accumulation value by one unit update value in response to a detected coincidence of stochastic bits streams of input vectors applied on an update row and update column control lines connected to the RPU device, generating a weight update control signal in response to the accumulated weight value reaching a predefined threshold value, and adjusting a conductance level of the tunable resistive device by one unit conductance value in response to the weight update control signal, wherein the one unit conductance value corresponds to one unit weight value of the RPU device.

Abstract: A system for classification using expert data includes at least a server. The system includes an expert submission processing module operating on the at least a server, the expert submission processing module designed and configured to receive at least an expert submission relating constitutional data to ameliorative recommendation data. The system includes a model generator operating on the at least a server, the model generator designed and configured to generate, using the at least an expert submission, and a constitutional inquiry, an ameliorative output. The system includes a client-interface module operating on the at least a server, the client-interface module designed and configured to receive, from a user client device, the constitutional inquiry and transmit, to the user client device, the ameliorative output.

Abstract: Systems and methods for neural network processing are provided. A method in a system comprising a plurality of nodes interconnected via a network, where each node includes a plurality of on-chip memory blocks and a plurality of compute units, is provided. The method includes upon service activation receiving an N by M matrix of coefficients corresponding to the neural network model. The method includes loading the coefficients corresponding to the neural network model into the plurality of the on-chip memory blocks for processing by the plurality of compute units. The method includes regardless of a utilization of the plurality of the on-chip memory blocks as part of an evaluation of the neural network model, maintaining the coefficients corresponding to the neural network model in the plurality of the on-chip memory blocks until the service is interrupted or the neural network model is modified or replaced.

Abstract: Methods and systems may provide for technology to generate a plurality of predictions associated with a test period based on an analytics model and generate an error model of the plurality of predictions based on first non-parametric time series data associated with the test period, wherein the error model is to be generated independently of the analytics model. Additionally, the technology may automatically determine an interval based on the error model, the first non-parametric time series data and a margin of error input, wherein the interval is to include an upper bound on a scale associated with the first non-parametric time series data at a time instance in a forecast period and a lower bound on the scale at the time instance.

Abstract: A method, system and computer-usable medium are disclosed for the use of a text simplification in a question answering (QA) system to improve ingestion quality and QA performance. A source corpus is processed to generate a parsed source corpus, which in turn is processed to generate a shadow corpus of simplified text. The parsed source corpus and the shadow corpus are then processed to generate derived data resources. A user query is processed to generate a set of merged candidate answer variants which are in turn processed to generate a corresponding converged feature vector representing each merged candidate answer variant. Feature values associated with each converged feature vector are then used to rank the merged candidate answer variants. A ranked set of merged candidate answer variants is then provided to the user.

Abstract: The present disclosure provides systems and methods to reduce computational costs associated with convolutional neural networks. In addition, the present disclosure provides a class of efficient models termed “MobileNets” for mobile and embedded vision applications. MobileNets are based on a straight-forward architecture that uses depthwise separable convolutions to build light weight deep neural networks. The present disclosure further provides two global hyper-parameters that efficiently trade-off between latency and accuracy. These hyper-parameters allow the entity building the model to select the appropriately sized model for the particular application based on the constraints of the problem. MobileNets and associated computational cost reduction techniques are effective across a wide range of applications and use cases.

Abstract: This disclosure describes methods, apparatuses, and systems for network training and testing for evaluating hardware characteristics and for hardware selection. For example, a sensor can capture a dataset, which may be transformed into a plurality of modified datasets to simulate changes to hardware. Each of the plurality of modified datasets may be used to individually train an untrained neural network, thereby producing a plurality of trained neural networks. In order to evaluate the trained neural networks, each neural network can be used to ingest an evaluation dataset to perform a variety of tasks, such as identifying various objects within the dataset. A performance of each neural network can be determined and compared. A performance curve can be determined for each characteristic under review, facilitating a selection of one or more hardware components and/or configurations.

Abstract: A prediction system provided with an integrated communications interface may include at least one processor configured to scrape the Internet to identify a currently pending legislative bill and information about legislators slated to vote on the pending bill. The processor may parse the information to determine a tendency position for each legislator. The processor may transmit for display to a system user a virtual whipboard that groups legislators into a plurality of groups based on determined tendency positions. The processor may receive a selected one of the plurality of groups of legislators for a communication interaction based on the determined tendency position of the group and access a legislator database that includes legislative communication addresses of legislative personnel scraped from the Internet and divided into a plurality of legislative function categories and receive from the system user a selection of at least one of the plurality of legislative function categories.

Abstract: A system and a computer program product are provided for evaluating question-answer pairs in an answer key by generating a predicted answer to a test question based on the answer key modification history for comparison matching against a generated answer that is generated in response to the test question, and then comparing the predicted answer and generated answer to determine an accuracy score match indication therebetween so as to present an indication that the answer key may have a problem if there is a match between the predicted answer and generated answer.

Abstract: A system and method for controlling a nodal network. The method includes estimating an effect on the objective caused by the existence or non-existence of a direct connection between a pair of nodes and changing a structure of the nodal network based at least in part on the estimate of the effect. A nodal network includes a strict partially ordered set, a weighted directed acyclic graph, an artificial neural network, and/or a layered feed-forward neural network.

Abstract: Aspects provide human detector devices based on neuronal response, wherein the devices are configured to obtain electroencephalogram signals from an entity during a presentation of first sensory information to the entity, and compares the obtained electroencephalogram signals to each of a plurality of trained electroencephalogram signal profile portions that are labeled as the first sensory information that represent electroencephalogram signals most commonly generated by different persons as a function of presentation to the persons of sensory information corresponding to the first sensory information. Thus, the configured processor determines whether the entity is a human as a function of a strength of match of the obtained electroencephalogram signals to ones of the trained electroencephalogram signal profile portions labeled as first sensory information that have highest most-common weightings.

Abstract: A data processing method in a data processing device is provided. First to-be-processed data input into a neural network are obtained. Iterative training is performed on the neural network for a first preset number of times by using first target data in the first to-be-processed data, to obtain a seed model of the neural network. First newly added data generated after an elapse of time corresponding to the first time window is obtained, and the first newly added data and the first to-be-processed data are combined into second to-be-processed data. Iterative training is performed on the seed model for a second preset number of times by using second target data in the second to-be-processed data, to obtain a first incremental model of the neural network. A first preset area overlaps between the second time window and the first time window. The first incremental model online is published.

Abstract: Data is classified using semi-supervised data. Sparse coefficients are computed using a decomposition of a Laplacian matrix. (B) Updated parameter values are computed for a dimensionality reduction method using the sparse coefficients, the Laplacian matrix, and a plurality of observation vectors. The updated parameter values include a robust estimator of a decomposition matrix determined from the decomposition of the Laplacian matrix. (B) is repeated until a convergence parameter value indicates the updated parameter values for the dimensionality reduction method have converged. A classification matrix is defined using the sparse coefficients and the robust estimator of the decomposition of the Laplacian matrix. The target variable value is determined for each observation vector based on the classification matrix.

Abstract: Systems and methods are disclosed that enable distributed execution of prediction models by disparate, remote systems. Prediction model code is transmitted to the disparate, distributed systems for execution by the disparate, remote systems. Default model input data may be independently modified by a given system, and the modified input data may be used when the given system executes the model. Model predictions and associated model parameters are received from the disparate, distributed systems. The accuracy of the received model predictions from the disparate, distributed systems are analyzed. Based on the analyzed accuracy of the received model predictions, a determination is made as to which model predictions satisfy at least a first criterion. Computer-based resources are allocated using the determination as to which model predictions satisfy at least the first criterion.

Abstract: Processors and methods for neural network processing are provided. A method in a processor including a pipeline having a matrix vector unit (MVU), a first multifunction unit connected to receive an input from the matrix vector unit, a second multifunction unit connected to receive an output from the first multifunction unit, and a third multifunction unit connected to receive an output from the second multifunction unit is provided. The method includes decoding a chain of instructions received via an input queue, where the chain of instructions comprises a first instruction that can only be processed by the matrix vector unit and a sequence of instructions that can only be processed by a multifunction unit. The method includes processing the first instruction using the MVU and processing each of instructions in the sequence of instructions depending upon a position of the each of instructions in the sequence of instructions.

Abstract: A method of configuring a hardware implementation of a Convolutional Neural Network (CNN), the method comprising: determining, for each of a plurality of layers of the CNN, a first number format for representing weight values in the layer based upon a distribution of weight values for the layer, the first number format comprising a first integer of a first predetermined bit-length and a first exponent value that is fixed for the layer; determining, for each of a plurality of layers of the CNN, a second number format for representing data values in the layer based upon a distribution of expected data values for the layer, the second number format comprising a second integer of a second predetermined bit-length and a second exponent value that is fixed for the layer; and storing the determined number formats for use in configuring the hardware implementation of a CNN.

Abstract: Presented herein are techniques for training a central/global machine learning model in a distributed machine learning system. In the data sampling techniques, a subset of the data obtained at the local sites is intelligently selected for transfer to the central site for use in training the central machine learning model. In the model merging techniques, distributed local training occurs in each local site and copies of the local machine learning models are sent to the central site for aggregation of learning by merging of the models. As a result, in accordance with the examples presented herein, a central machine learning model can be trained based on various representations/transformations of data seen at the local machine learning models, including sampled selections of data-label pairs, intermediate representation of training errors, or synthetic data-label pairs generated by models trained at various local sites.

Abstract: A system and a computer program product are provided for evaluating question-answer pairs in an answer key by generating a predicted answer to a test question based on the answer key modification history for comparison matching against a generated answer that is generated in response to the test question, and then comparing the predicted answer and generated answer to determine an accuracy score match indication therebetween so as to present an indication that the answer key may have a problem if there is a match between the predicted answer and generated answer.