Embodiments for ensemble policy generation for prediction systems by a processor. Policies are generated and/or derived for a set of ensemble models to predict a plurality of target variables for streaming data such that the plurality of policies enables dynamic adjustment of the prediction system. One or more of the policies are updated according to one or more error states of the set of ensemble models.

Certain aspects of the present disclosure provide techniques for adaptively reducing the bit size of features in a training data set used to train a machine learning model. An example method generally includes receiving a data set to be used in training a machine learning model and a definition of the machine learning model to be trained. A reduced number of bits to represent features in the data set is determined based on values of each feature in the data set and the definition of the machine learning model. A reduced bit-size data set is generated by reducing a bit size of each feature in the data set according to the reduced number of bits, and the reduced bit-size data set is encrypted using a homomorphic encryption scheme. A machine learning model is trained based on the encrypted reduced bit-size data set.

In an example, a computer-implemented method to structure an analytical workflow that improves reasoning based on a problem context and demonstrated abilities of each individual user may include displaying a reasoning problem to an analyst. The method may include receiving input from the analyst to identify a reasoning problem type of the reasoning problem. providing a recommended analytic approach for the reasoning problem type to the analyst. The method may include assisting the analyst in analyzing and evaluating one or more information sources relevant to the reasoning problem. The method may include guiding the analyst through a structured technique (ST) to support reasoning of the analyst in formulation of a solution to the reasoning problem. The method may include generating a report that includes the analyst's solution to the reasoning problem based on input from the analyst.

A machine learning device that searches for a first parameter of a component of a servo control device that controls a servo motor includes: a solution detection unit that acquires a set of evaluation function values used during machine learning or after machine learning, plots the set of evaluation function values in a search range of the first parameter or a second parameter used for searching for the first parameter, and detects whether a search solution is at an edge of the search range or is in a predetermined range from the edge; and a range changing unit that changes the search range to a new search range of the first parameter or the second parameter based on the estimation made on evaluation function values of an evaluation function expression when the search solution is at the edge of the search range or is in the predetermined range.

An apparatus in one embodiment comprises at least one processing platform including a plurality of processing devices. The processing platform is configured to receive a request to execute one or more predictive models for generating a delivery date, to initiate execution of the one or more predictive models responsive to the request, and to invoke a plurality of machine learning algorithms using data from a plurality of data sources when executing the one or more predictive models. The processing platform is further configured to capture the data from the plurality of data sources and organize the data into a sparse matrix, to automatically generate the delivery date, and to automatically transmit the delivery date to one or more user devices.

Embodiments of this application disclose a vehicle-based data processing method performed by a computer device. The method includes: determining at least two predicted offsets of a first vehicle, a first traveling state of the first vehicle, and a second traveling state of a second vehicle; determining, according to the first traveling state and the second traveling state, first lane change payoffs of the predicted offsets when the second vehicle is in a yielding prediction state, and determining second lane change payoffs when the second vehicle is in a non-yielding prediction state; and determining a predicted yielding probability of the second vehicle, generating target lane change payoffs of the predicted offsets according to the predicted yielding probability and the first lane change payoffs and the second lane change payoffs of the predicted offsets, and determining a predicted offset having a maximum target lane change payoff as a target predicted offset.

In aspects of the present disclosure, a circuit interrupter includes a housing, a conductive path, a switch which selectively interrupts the conductive path, sensor(s), memory, and a controller within the housing. The sensor(s) measure electrical characteristic(s) of the conductive path. The memory stores an arc detection program that implements a machine learning model and includes a field-updatable program portion and a non-field-updatable program portion, where the field-updatable program portion includes program parameters used by the non-field-updatable program portion to decide between presence or absence of an arc fault. The controller executes the arc detection program to compute input data for the machine learning model based on the sensor measurements, decide between presence of an arc event or absence of an arc event based on the input data, and cause the switch to interrupt the conductive path when the decision indicates presence of an arc event.

One embodiment provides a method, including: producing, for each of a plurality of containers, a resource profile for each thread in each of the plurality of containers; identifying, for each of the plurality of containers and from, at least in part, the resource profiles, container dependencies between threads on a single of the plurality of containers; determining service dependencies between threads across different of the plurality of containers; scheduling, based upon the container dependencies and the service dependencies, threads to cores, wherein the scheduling is based upon minimizing thread processing times; and publishing the container dependencies and the service dependencies on a registry of the node clusters.

Facilitating striated random restart during routing of resources using a heuristic approach is provided herein. Operations of a system can include, separating objects of a group of objects into a first grouping and a second grouping, assigning first objects of the first grouping and second objects of the second grouping to respective resource elements, resulting in a first classification. A search space is enabled and includes, at respective iterations of a group of iterations, selecting a defined priority level, removing third objects of the second grouping that do not satisfy the defined priority level from the first classification, and assigning the third objects to the respective resource elements, resulting in a second classification. The first objects and the second objects, other than the third objects, remain assigned according to the first classification. According to an implementation, the schedule search reduces a search space of the group of objects.

Systems and methods to semantically compare product configuration models. A method includes receiving a first configuration model and a second configuration model. The method includes generating a first order logic (FOL) representation of the first configuration model and an FOL representation of the second configuration model. The method includes performing a satisfiability modulo theories (SMT) solve for nonequivalence satisfiability on the FOL representation of the first configuration model and the FOL representation of the second configuration model. The method includes storing an indication that the first configuration model is equivalent to the second configuration model when the SMT solve for nonequivalence satisfiability is not satisfied.