Abstract: Systems and methods are provided for reducing failure rates of a manufactured products. Manufactured products may be clustered together according to similarities in their production data. Manufactured product clusters may be analyzed to determine mechanisms for failure rate reduction, including adjustments to test quality parameters, product formulas, and product processes. Recommended product adjustments may be provided.

Abstract: A device, system, and method is provided to mimic a pre-trained target model without access to the pre-trained target model or its original training dataset. A set of random or semi-random input data may be sent to randomly probe the pre-trained target model at a remote device. A set of corresponding output data may be received from the remote device that is generated by applying the pre-trained target model to the set of random or semi-random input data. A random probe training dataset may be generated comprising the set of random or semi-random input data and corresponding output data generated by randomly probing the pre-trained target model. A new model may be trained with the random probe training dataset so that the new model generates substantially the same corresponding output data in response to said input data to mimic the pre-trained target model.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for organizing trained and untrained neural networks. In one aspect, a neural network device includes a collection of node assemblies interconnected by between-assembly links, each node assembly itself comprising a network of nodes interconnected by a plurality of within-assembly links, wherein each of the between-assembly links and the within-assembly links have an associated weight, each weight embodying a strength of connection between the nodes joined by the associated link, the nodes within each assembly being more likely to be connected to other nodes within that assembly than to be connected to nodes within others of the node assemblies.

Abstract: The quantum circuit learning device includes a signal input unit that provides a quantum circuit including plural quantum bits with an input signal, a signal acquisition unit that observes states of quantum bits that the quantum circuit includes and acquires an output signal based on the observed states, and an adjustment unit that adjusts a circuit parameter that defines a circuit configuration of the quantum circuit, using an output signal that the signal acquisition unit acquires and a cost function that is set based on a teacher signal corresponding to the output signal.

Abstract: Disclosed herein is a technique for implementing a framework that enables application developers to enhance their applications with dynamic adjustment capabilities. Specifically, the framework, when utilized by an application on a mobile computing device that implements the framework, can enable the application to establish predictive models that can be used to identify meaningful behavioral patterns of an individual who uses the application. In turn, the predictive models can be used to preempt the individual's actions and provide an enhanced overall user experience. The framework is configured to interface with other software entities on the mobile computing device that conduct various analyses to identify appropriate times for the application to manage and update its predictive models. Such appropriate times can include, for example, identified periods of time where the individual is not operating the mobile computing device, as well as recognized conditions where power consumption is not a concern.

Abstract: Various embodiments of methods and systems, including computer programs encoded on computer storage media described herein are directed to real-time situation determination based on distributed event data. According to various embodiments, the system receives event data from one or more computing devices. The system provides a machine learning model configured to use a plurality of interconnected check-point evaluators to evaluate the received event data and determine an occurrence of a situation. The system evaluates event values, via one or more check-point evaluator of the plurality of interconnected check-point evaluators, whether the event values meet criteria for one or more situation indicators. Based on the evaluation of the event values the system determines the occurrence of the situation.

Abstract: Systems and methods are provided for performing multi-objective optimizations with a relatively large number of objectives to which optimization is to be performed. The objectives of the optimization problem may be partitioned to two or more subsets (e.g., overlapping or non-overlapping subsets) of objectives, and partial optimization(s) may be performed using a subset or combination of subsets of the objectives. One or more of the partial optimizations may use one or more pareto-optimized chromosomes from a prior partial optimization. A final full optimization may be performed according to all of the objectives of the optimization problem and may use one or more chromosomes of any preceding partial optimization as a starting point for finding a final solution to the optimization problem. Any variety of processes may be employed to mitigate archive explosion that may be associated with relatively large objective sets.

Abstract: Dynamically updating, or retraining and updating, artificial intelligence (AI)/machine learning (ML) models in digital processes at runtime is disclosed. Production operation may not need to be stopped for AI/ML model update or retraining and update. The update steps and/or retraining steps for the AI/ML model may be included as part of the digital process. The AI/ML model update may be requested from internal logic (e.g., from the evaluation of a condition, by an that expression calls for the AI/ML model, etc.), external requests (e.g., from external triggers in a finite state machine (FSM), such as a file change, database data, a service call, etc.), or both. Automation of AI/ML model updates or retraining and updates may be provided, where the software reloads/reinitializes/re-instantiates with a retrained and/or updated AI/ML model after (and possibly immediately after) the AI/ML model becomes available.

Abstract: Systems, methods, and computer-readable media are provided for facilitating clinical decision making by directing the emission of computer-generated health-care related recommendations towards contexts in which the recipient will likely find the recommendations salient and will likely welcome them and act upon them. ‘Uptake’ of computer-generated recommendations for diagnostic tests or therapeutic interventions is thereby substantially increased, and ‘alert fatigue’ is substantially decreased. Embodiments of our technology overcome certain drawbacks associated with the prior art by providing a means for ascertaining which decision-support recommendations are likely to be favorably considered by the recipient and acted-upon (recommendation ‘uptake’). System and method embodiments for providing a predicted probability of user uptake of a context-specific system-generated recommendation patient are disclosed herein and for applying that information to decide whether or not to emit the relevant recommendation.

Abstract: A method, system and computer-usable medium for performing cognitive computing operations comprising receiving streams of data from a plurality of data sources; processing the streams of data from the plurality of data sources, the processing the streams of data from the plurality of data sources performing data enriching for incorporation into a cognitive graph; defining a cognitive persona within the cognitive graph, the cognitive persona corresponding to an archetype user model, the cognitive persona comprising a set of nodes in the cognitive graph; associating a user with the cognitive persona; and, performing a cognitive computing operation based upon the cognitive persona associated with the user.

Abstract: The present disclosure relates to processing support data to increase a self-support knowledge base. In some embodiments, assisted support data is received comprising a record of an interaction between a user and a support professional. In certain embodiments, a support data set is extracted from the assisted support data. In some embodiments, feedback related to the support data set is received. The feedback may include an indication that the support data set is ready to be included in the self-support knowledge base. In some embodiments, upon determining, based on the feedback, that the support data set is ready to be used for self-support, the support data set is added to the self-support knowledge base. The self-support knowledge base may be accessible by a plurality of users.

Abstract: Embodiments of a method, apparatus, device and computer readable storage medium for classifying driving scenario data includes: acquiring a first driving scenario data set from a crowdsourcing platform, driving scenario data in the first driving scenario data set having been classified; generating a driving scenario classification model at least based on the first driving scenario data set, for classifying driving scenario data collected by a collection entity; acquiring a rule for classifying the driving scenario data, the rule is generated based on a result of classifying the driving scenario data collected by the collection entity using the driving scenario classification model; updating the driving scenario classification model at least based on the rule.

Abstract: Pattern classification systems and methods are disclosed. The pattern classification systems and methods employ one or more classification networks that can parse multiple patterns simultaneously while providing a continuous feedback about its progress. Pre-synaptic inhibition is employed to inhibit feedback connections to permit more flexible processing. Various additional improvements result in highly robust pattern recognition systems and methods that are suitable for use in research, development, and production.

Abstract: Techniques are described automatically determining runtime configurations used to execute recurrent neural networks (RNNs) for training or inference. One such configuration involves determining whether to execute an RNN in a looped, or “rolled,” execution pattern or in a non-looped, or “unrolled,” execution pattern. Execution of an RNN using a rolled execution pattern generally consumes less memory resources than execution using an unrolled execution pattern, whereas execution of an RNN using an unrolled execution pattern typically executes faster. The configuration choice thus involves a time-memory tradeoff that can significantly affect the performance of the RNN execution. This determination is made automatically by a machine learning (ML) runtime by analyzing various factors such as, for example, a type of RNN being executed, the network structure of the RNN, characteristics of the input data to the RNN, an amount of computing resources available, and so forth.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing a layer output for a convolutional neural network layer, the method comprising: receiving a plurality of activation inputs; forming a plurality of vector inputs from the plurality of activation inputs, each vector input comprising values from a distinct region within the multi-dimensional matrix; sending the plurality of vector inputs to one or more cells along a first dimension of the systolic array; generating a plurality of rotated kernel structures from each of the plurality of kernel; sending each kernel structure and each rotated kernel structure to one or more cells along a second dimension of the systolic array; causing the systolic array to generate an accumulated output based on the plurality of value inputs and the plurality of kernels; and generating the layer output from the accumulated output.

Abstract: Systems and techniques for hierarchical tournament-based machine learning predictions are described herein. A machine learning selection model may be trained with training data. A configuration may be received that includes the metric and a target prediction. A set of evaluation component combinations may be selected using the machine learning selection model. Each evaluation component combination of the set of evaluation component combinations may include an algorithm, a hierarchical learning model corresponding to a level of a hierarchy, and a prediction model for the target prediction. The set of evaluation component combinations may be transmitted to a cluster of computing nodes. Output results may be received for the set of evaluation component combinations. The output results may be evaluated using the metric to determine a winning evaluation component combination. The winning evaluation component combination may be stored in storage for use in calculating future predictions for the target prediction.

Abstract: A neural network runs a known input data set using an error free power setting and using an error prone power setting. The differences in the outputs of the neural network using the two different power settings determine a high level error rate associated with the output of the neural network using the error prone power setting. If the high level error rate is excessive, the error prone power setting is adjusted to reduce errors by changing voltage and/or clock frequency utilized by the neural network system. If the high level error rate is within bounds, the error prone power setting can remain allowing the neural network to operate with an acceptable error tolerance and improved efficiency. The error tolerance can be specified by the neural network application.

Abstract: A supervised learning processing (SLP) system and method provide cooperative operation of a network of supervised learning processors to concurrently distribute supervised learning processor training, generate predictions, provide prediction driven responses to input objects, and provide operational sequencing to concurrently control and distribute supervised learning processor training and provide predictive responses to input data. The SLP system can dynamically sequence SLP subsystem operations to improve resource utilization, training quality, and/or processing speed. A system monitor-controller can dynamically determine if process environmental data indicates initiation of dynamic subsystem processing sequencing. Concurrently training SLP's provides accurate predictions of input objects and responses thereto and enhances the network by providing high quality value predictions and responses and avoiding potential training and operational delays.

Abstract: In various embodiments, a process for obtaining a generated dataset with a predetermined bias for evaluating algorithmic fairness of a machine learning model includes receiving an input dataset and generating an anonymized reconstructed dataset based at least on the input dataset. The process includes introducing a predetermined bias into the generated dataset, forming an evaluation dataset based at least on the generated dataset with the predetermined bias, and outputting the evaluation dataset. In various embodiments, a process for training a generative model includes configuring a generative model and receiving training data, where the training data includes a tabular dataset. The process includes using computer processor(s) and the received training data to train the generative model, where the generative model is sampled to generate a dataset with a predetermined bias.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying user behavior as anomalous. One of the methods includes obtaining user behavior data representing behavior of a user in a subject system. An initial model is generated from training data, the initial model having first characteristic features of the training data. A resampling model is generated from the training data and from multiple instances of the first representation for a test time period. A difference between the initial model and the resampling model is computed. The user behavior in the test time period is classified as anomalous based on the difference between the initial model and the resampling model.