Abstract: A device extracts data files from different data sources, and processes the data files to generate data from the data files. The device trains a first artificial intelligence model and a second artificial intelligence model based on historical data parameters, wherein the historical data parameters are associated with historical data previously processed by the device. The device utilizes the trained first artificial intelligence model and the data to generate data parameters for the data, and utilizes the trained second artificial intelligence model, the data, and the data parameters to generate a data structure configuration. The device generates a data structure for the data based on the data parameters and the data structure configuration, stores the data in the data structure, and provides a user interface for accessing the data structure.

Abstract: Systems and methods are disclosed for improving Physical Downlink Control Channel (PDCCH) allocation for improved downlink scheduling. In some embodiments, a method of operation of a network node in a cellular communications network comprises transmitting a downlink assignment to a wireless device, the downlink assignment comprising an indication of whether the downlink assignment applies to a single Transmit Time Interval (TTI) or multiple consecutive TTIs, and transmitting a downlink data transmission to the wireless device during a single TTI or multiple downlink data transmissions to the wireless device during multiple consecutive TTIs according to the downlink assignment. In this manner, the network node does not need to transmit, and the wireless device does not need to receive/decode, a separate downlink assignment in each TTI. This may be beneficial to, e.g., mitigate PDCCH congestion.

Abstract: Computer vision systems and methods for machine learning using image hallucinations are provided. The system generates image hallucinations that are subsequently used to train a deep neural network to match image patches. In this scenario, the synthesized changes serve in the learning of feature-embedding that captures how a patch of an image might look like from a different vantage point. In addition, a curricular learning framework is provided which is used to automatically train the neural network to progressively learn more invariant representations.

Abstract: An example of an apparatus including a network interface to receive data about a data application via an external network is provided. The apparatus further includes a collection engine in communication with the network interface to receive the data. The apparatus also includes a metadata generation engine to generate application metadata associated with the data application. The apparatus includes a memory storage unit to maintain a database of metadata. The database includes the application metadata associated with the data application. The apparatus also includes a processor to apply the database of metadata to a lookup table.

Abstract: An example system includes memory to store a metric database table and a processor coupled to the memory. The metric database table includes a metric key and metric logic associated with the metric key. The metric logic is executable against a device datastore to extract metric data from the device datastore. The device datastore is to store operational data of a plurality of electronic devices. The processor is to query the metric database table to obtain the metric logic. The processor is to execute the metric logic against the device datastore to extract the metric data. The processor is to output extracted metric data in association with the metric key to an incident datastore to monitor for incidents related to operations of the plurality of electronic devices.

Abstract: A system is provided including a memory in communication with a processor. The memory is to store a value of an attribute. The processor is to obtain from a source data structure the value. The processor is also to store in a further data structure the value in association with a descriptor identifying the source data structure. Moreover, the processor is to store in the further data structure an identifier in association with the value. The identifier is incrementable and unique to the value among values associated with the descriptor. In addition, the processor is to output the further data structure.

Abstract: Systems and methods for unsupervised representation learning by sorting sequences are provided. An unsupervised representation learning approach is provided which uses videos without semantic labels. The temporal coherence as a supervisory signal can be leveraged by formulating representation learning as a sequence sorting task. A plurality of temporally shuffled frames (i.e., in non-chronological order) can be used as inputs and a convolutional neural network can be trained to sort the shuffled sequences and to facilitate machine learning of features by the convolutional neural network. Features are extracted from all frame pairs and aggregated to predict the correct sequence order. As sorting shuffled image sequence requires an understanding of the statistical temporal structure of images, training with such a proxy task can allow a computer to learn rich and generalizable visual representations from digital images.

Abstract: A method includes creating, by one or more computing devices, a first pool of memory resources from a plurality of memory resources in a cluster. The first pool comprises memory resources that (1) have been allocated to a respective guest virtual machine that is not currently being used and (2) that have been marked as unused by the respective guest virtual machine. The method also includes invoking, by the one or more computing devices, a number of operations on the memory resources in the first pool, wherein the number of operations is designed to signal a memory resource manager to reclaim the allocated memory resources that have been marked as unused from the first pool into an unallocated pool of memory resources in the cluster.

Abstract: A processor-implemented battery management method includes iteratively determining fractions of components in an electrode of a battery; calculating a composite state of charge (SoC) of the battery by applying the iteratively determined fractions of the components to a composite cathode model used to predict a state of the battery; and predicting the state of the battery based on the calculated composite SoC.

Abstract: Methods and apparatuses are described for system for identifying and mitigating high-risk database queries through ranked variance analysis. A server identifies database queries executed against databases in a production computing environment within a predetermined time period, each database query associated with execution plans and each execution plan having corresponding plan data elements. For each database query: the server generates execution variance data for the execution plans for a database query based upon the corresponding plan data elements, comprising: determining an execution time variance between the execution plans; and determining a buffer gets variance between the execution plans. The server ranks the database queries according to (i) the execution time variance, and (ii) the buffer gets variance.

Abstract: Processes and apparatus for isomerizing hydrocarbons are provided. The process comprises providing a feed stream comprising C5, C6 and C7+ hydrocarbons. The feed stream is separated to produce a first stream rich in C5 and C6 hydrocarbons and a second stream rich in benzene and C7+ hydrocarbons. The first stream is isomerized in the presence of isomerization catalyst and hydrogen in an isomerization zone under isomerization conditions to produce an isomerized stream. The second stream is contacted with a benzene saturation catalyst at benzene saturation conditions to produce a saturation effluent stream comprising cyclohexane.

Abstract: The present disclosure relates to a fuel-air premixer for a turbine system. The fuel-air premixer includes a swirler and a centerbody. The swirler is configured to direct a flow of air through the premixer, and the centerbody is configured to inject fuel into the flow of air. Additionally, the centerbody includes an airfoil shape that reduces and/or substantially eliminates recirculation pockets to prevent autoignition and/or flame holding in a combustion chamber. Accordingly, the turbine system may produce fewer NOx emissions.

Abstract: A cooling system includes a charge air cooler system that includes a first stage that receives charge air via a charge air flow path and receives coolant fluid via a first coolant fluid flow path. A second stage receives charge air from the first stage via the charge air flow path, outputs the charge air, and receives the coolant fluid via a second coolant fluid flow path. A third stage receives and outputs the charge air from the second stage via the charge air flow path and receives the coolant fluid via a third coolant fluid flow path. The cooling system includes a low temperature radiator system that includes a low temperature radiator that directs the coolant fluid toward the third stage via the third coolant fluid flow path and includes a high temperature radiator system that directs the coolant fluid toward the first stage and second stage.

Abstract: Methods, compositions and kits for regulating complement activity or treating a complement activity disorder in a subject using soluble, membrane-independent CD59 protein, methods of assaying human macular degeneration (MD), and methods and kits for assaying potential therapeutic agents for treatment of human MD are provided herein.

Abstract: A system and method includes operation of a generation network to generate first generated computed tomography data based on a first instance of surface data, determination of a generation loss based on the first generated computed tomography data and on a first instance of computed tomography data which corresponds to the first instance of surface data, operation of a discriminator network to discriminate between the first generated computed tomography data and the first instance of computed tomography data, determination of a discriminator loss based on the discrimination between the first generated computed tomography data and the first instance of computed tomography data, determination of discriminator gradients of the discriminator network based on the discriminator loss, and updating of weights of the generation network based on the generation loss and the discriminator gradients.

Abstract: Live video streams are produced using a network server system. The network server system initially receives, via a network, one or more captured video streams that are live streams captured from a remotely-located camera, phone or the like. The captured streams are forwarded to a control device via the network to thereby permit a user of the control device to select one of the captured video streams for output to the video production stream. In response to a command received from the control device that indicates the selected capture stream, the selected capture video streams is encoded for output as the video production stream.

Abstract: A method for training a learning-based medical scanner including (a) obtaining training data from demonstrations of scanning sequences, and (b) learning the medical scanner's control policies using deep reinforcement learning framework based on the training data.

Abstract: A method for providing obstacle alerts to an in-flight aircraft has been developed. First, parameters of an in-flight aircraft are transmitted to a ground based processor station. The station calculates an aircraft safety envelope based on these parameters. The station accesses the characteristics of obstacles stored in a terrain database and calculates an obstacle safety envelope. Finally, the station determines if the aircraft safety envelope conflicts with the obstacle safety envelope and generates an alert for the aircraft if a conflict exists.

Abstract: Methods, systems, and products translate addresses in networks. A residential gateway translates requests for content such that the residential gateway appears as both a requestor and a destination for requested content, regardless of an actual requesting device or a desired output device.

Abstract: An event processing system for distributing geospatial computations with geo-fences is disclosed. The system can partition and distribute geo-fences efficiently for parallel computation, and can track a potentially unlimited number of moving objects. A computing device determines a minimum bounding rectangle (MBR) covering a geographic area indexed by a geo-grid, and partitions the MBR and a set of geo-fences to a plurality of processing nodes. The computing device receives input events of a stream comprising locations within the geographic area. The device distributes the events to partitions on processing nodes, based at least on the events' locations. The device can determine a partition identifier corresponding to a respective location in constant time. The computing device then combines partition results to obtain a single geospatial computation result.