Abstract: A method is provided. The method includes determining an open circuit voltage of a battery of a vehicle. The method also includes determining a voltage of a current provided to the battery of the vehicle. The method further includes determining a impedance indicator based on the voltage, the open circuit voltage, and the current provided to the battery of the vehicle. The method further includes determining a health of the battery based on the impedance indicator.

Abstract: Persistent storage contains an original document and a plurality of summaries of the original document produced by summarization models. One or more processors: provide, to an entity extractor, the original document; receive, from the entity extractor, a list of entities within the original document; provide, to a query generator, the original document and the list of entities; receive, from the query generator, a set of queries answerable by the original document; provide, to a query answerer, the set of queries, the original document, and the plurality of summaries; receive, from the query answerer and for the set of queries, a set of document answers corresponding to the original document and sets of summary answers corresponding to the plurality of summaries; provide, to an answer matcher, the set of document answers and the sets of summary answers; and receive, from the answer matcher, scores for the plurality of summaries.

Abstract: This disclosure relates to a fully software-defined, fully virtualized, and customizable mobile communication platform deployed on public cloud infrastructure. Such mobile networks allows for end to end control of automatic and programmatic deployment and configuration of the mobile network components. The implementations below effectively enables instant creation and deployment of a true private global end-to-end Software Defined Network (SDN) for 3G, 4G, LTE, and 5G mobile communication from the ground up. Users will effectively act their own mobile carrier, allowing them to customize the features available to them via a programmatic interface.

Abstract: A wireless user device wirelessly receives an indication of selected resource blocks in a frequency channel that are selected for radio signal measurement and unselected resource blocks in the frequency channel that are unselected for the radio signal measurement. The wireless user device wirelessly measures a radio signal metric for the selected resource blocks in the radio channel and does not measure the radio signal metric for the unselected resource blocks in the radio channel. The wireless user device wirelessly transfers the radio signal metric for the selected resource blocks in the radio channel. The wireless user device wirelessly receives a schedule for the selected resource blocks in the radio channel. The wireless user device wirelessly receives data over the selected resource blocks in the radio channel based on the schedule.

Abstract: An OCR filter described herein filters non-textual files in scanned customer data from optical character recognition (OCR) and pattern analysis of text generated thereof for sensitive customer data. The OCR filter is trained on files labelled using feature values for features generated from OCR applied to the corresponding files. Moreover, the OCR filter stores internal representations of the files during training to avoid leaking potential sensitive customer data contained therein. Once trained, performance of the OCR filter in filtering files comprising image data without text is evaluated according to false positive rates and false negative rates by comparing classifications of the OCR filter to classifications according to feature values for features generated from OCR. Evaluation of the OCR filter ensures continued model performance and informs model updates.

Abstract: A computing device is provided, including a logic subsystem with one or more processors, and memory storing instructions executable by the logic subsystem. These instructions are executed to obtain one or more source images, segment the one or more source images to generate a plurality of segments, determine a priority order for the plurality of segments, and transmit the plurality of segments to a remote computing device in the priority order. The plurality of segments are spatial components generated by spatial decomposition of the one or more source images and/or frequency components that are generated by frequency decomposition of the one or more source images. A remote computing device may receive these components in priority order, and perform certain algorithms on individual components without waiting for the entire image to upload.

Abstract: This disclosure provides a machine learning technique to predict a protein characteristic. A first training set is created that includes, for multiple proteins, a target feature, protein sequences, and other information about the proteins. A first machine learning model is trained and then used to identify which of the features are relevant as determined by feature importance or causal relationships to the target feature. A second training set is created with only the relevant features. Embeddings generated from the protein sequences are also added to the second training set. The second training set is used to train a second machine learning model. The first and second machine learning models may be any type of regressors. Once trained, the second machine learning model is used to predict a value for the target feature for an uncharacterized protein. The model of this disclosure provides 91% accuracy in predicting an ideal digestibility score.

Abstract: A technique for executing commands for an accelerated processing device is provided. The technique includes obtaining an iteration number and predication data from metadata for an iterative indirect command buffer; for each iteration indicated by the iteration number, performing commands of the iterative indirect command buffer as specified by the predication data; and ending processing of the iterative indirect command buffer in response to processing a number of iterations equal to the iteration number.

Abstract: The disclosure relates to sidechain functionalized organosiloxane compounds comprising polyalkyleneoxide (POA) and/or polyalkyleneoxide coupled zwitterionic moieties with various other modular side chains including reactive and or non-reactive groups. The present disclosure further also pertains to ice-phobic polymer formulations capable of curing on a substrate to form a surface that is ice-phobic, resistant to ice formation, and/or resistant to ice adhesion. The disclosed ice-phobic polymer formulations comprise one or more disclosed sidechain functionalized organosiloxane compound, one or more binder resin, and/or one or more lubricating liquid. The present disclosure further relates to articles comprising the disclosed ice-phobic polymer compositions, thereby providing articles having ice-phobic properties. This abstract is intended as a scanning tool for purposes of searching in the particular art and is not intended to be limiting of the present disclosure.

Abstract: The segment analysis system analyzes survey data to determine the influence each custom question/response combination (segment) has on a given aggregate scored survey metric for a given date/date range. The system removes from consideration all surveys that do not include a scored survey metric and date that matche the aggregate scored survey metric and given date/date range. The system further removes from consideration all surveys not pertaining received user-defined filtering. Once the system has eliminated all extraneous surveys from consideration, the system segments each question/response combinations across the pool of surveys to generate an influence score for each question/response combination. The system identifies which segment has the greatest positive and negative influence on the aggregate scored survey metric for the given date/date range. The system generates reports for the segment analysis and stores all segment analyses for further comparative analysis.

Abstract: Data from data sources may be processed at an edge device. The edge device may generate a local processing result, filter the data, and/or prioritize the data. Accordingly, data is transmitted from the edge device to the data platform, where it may be processed further. For example, a local processing result may be processed at the data platform, such that processing is performed without all of the data source data. In examples, at least a part of such data may remain at an edge device. The edge device may maintain a manifest of data stored by the edge device. The data platform may generate an aggregated manifest using manifests from associated edge devices, such that it may be determined where data is stored. As a result, the data platform may redirect requests to an associated edge device when it is determined that requested data is remote from the data platform.

Abstract: Disclosed are various examples for kernel level application data protection. In one example, a security label and a list of permitted applications are received. The security label is utilized to limit access to files that embed the security label. A security label map is written within a kernel layer of the client device. The security label map includes the security label and the list of permitted applications. A secured file is generated by embedding the security label within a file stored on the client device.

Abstract: Embodiments of the invention are directed to a system, method, or computer program product for multimodal and distributed database system structured for dynamic latency reduction. In this regard, the invention comprises a unified data layer structured to map a plurality of data storage mechanisms to a common abstraction and a query engine structured for heterogenous domain based data extraction without requiring input of schema-based queries. In some embodiments, the invention comprises determining (i) one or more data components and (ii) one or more associated data domains associated with the first domain-based query by parsing the user input based on derived metadata from data dictionaries associated with a unified data layer system component. Moreover, the invention is configured to extract stored data from each of a plurality of databases based on the associated one or more data domains.

Abstract: A method of tiled rendering is provided which comprises dividing a frame to be rendered, into a plurality of tiles, receiving commands to execute a plurality of subpasses of the tiles and interleaving execution of same subpasses of multiple tiles of the frame. Interleaving execution of same subpasses of multiple tiles comprises executing a previously ordered first subpass of a second tile between execution of the previously ordered first subpass of a first tile and execution of a subsequently ordered second subpass of the first tile. The interleaving is performed, for example, by executing the plurality of subpasses in an order different from the order in which the commands to execute the plurality of subpasses are stored and issued. Alternatively, interleaving is performed by executing one or more subpasses as skip operations such that the plurality of subpasses are executed in the same order.

Abstract: Systems and methods for authoring workflows for processing data from a large-scale dataset include defining a metadata schema for the large-scale dataset, and receiving user input defining a workflow as a plurality of operations to be performed on the data. Each of the operations includes input metadata formatted according to the metadata schema. The input metadata describes input data to be processed by the operation and identifying a location for the input data in the data storage system, programmed instructions for performing an atomic operation on the input data to generate output data; and output metadata formatted according to the metadata schema. The output metadata describes the output data and identifying a location for the output data in the data storage system.

Abstract: Multi-component modules (MCMs), including configurable electromagnetic interference (EMI) shield structures and related methods, are disclosed. An EMI shield enclosing an IC or another electrical component in an MCM can protect other components within the MCM from EMI generated by the enclosed component. The EMI shield also protects the enclosed component from the EMI generated by other electrical components. An EMI shield with sidewall structures, in which vertical conductors supported by a wall medium electrically couple a lid of the EMI shield to a ground layer in a substrate, provides configurable EMI protection in an MCM. The EMI shield may also be employed to increase heat dissipation. The sidewall structures of the EMI shield are disposed on one or more sides of an electrical component and are configurable to provide a desired level of EMI isolation.

Abstract: A system includes one or more processors and software, executable by the one or more processors that, when executed by the one or more processors, cause the system to retrieve Environmental, Social & Governance (ESG) related data from one or more first databases, retrieve financial-related data from one or more second databases, and generate a dynamically interactive graphical user interface (GUI), containing a combination of ESG and financial information, the GUI having multiple GUI Modules including a Screening GUI Module, a Ratings GUI Module, a Climate GUI Module, and a Sustainability GUI Module.

Abstract: A processor includes a task scheduling unit and a compute unit coupled to the task scheduling unit. The task scheduling unit performs a task dependency assessment of a task dependency graph and task data requirements that correspond to each task of the plurality of tasks. Based on the task dependency assessment, the task scheduling unit schedules a first task of the plurality of tasks and a second proxy object of a plurality of proxy objects specified by the task data requirements such that a memory transfer of the second proxy object of the plurality of proxy objects occurs while the first task is being executed.

Abstract: Memory cell sensing by charge sharing between two sense nodes is disclosed. A first sense node and a second sense node are pre-charged and the second node is discharged initiating charge sharing between the first sense node and the second sense node that results in an improved sense margin. Sensing circuitry disclosed herein may include one or more pre-charge circuits, sense enable circuits, and charge-sharing circuits. The increased sense margin achieved by sensing circuitry disclosed herein provides better noise immunity and more accurate sensing results.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods that utilize transformation and projection layers to implement neural networks in tabular data environments. For example, the disclosed systems identify tabular data sets and segregate measured tabular data values and placeholder tabular data values. In one or more embodiments, the disclosed systems transform the measured tabular data values to a neural network value range based on the distribution of the measured tabular data values. Moreover, the disclosed systems replace placeholder tabular data values with a constant value within the neural network value range. In addition, the disclosed systems identify numerical values and utilize a projection layer to generate feature vectors within a high-dimensionality feature space. In one or more embodiments, the disclosed systems utilize the resulting high-dimensionality tabular data set with a neural network to generate prediction results (e.g.