Abstract: Aspects of the disclosure are directed to a computation unit implementing a systolic array and configured for detecting errors while processing data on the systolic array. Checksum circuit in communication with a systolic array is configured to compute checksums and perform error detection while the systolic array processes input data. Instead of pre-generating checksums in input matrices, input matrices can be directly fed into the systolic array through the checksum circuit. On the output side, the checksum circuit can generate and compare checksums with checksums in an output matrix generated by the systolic array. Error checking the operations to generate the output matrix can be performed without delaying the operations of the systolic array, and without preprocessing the input matrices.

Abstract: Systems, methods and apparatus for multi-realm, computer-generated reality systems are disclosed. In an aspect, a multi-realm, computer-generated reality system may admit a plurality of participants to a session conducted in a managed multi-realm reality system, wherein each of the plurality of participants is admitted through a virtual reality context, extended reality context or an augmented reality context. The system may receive information generated by one or more sensors associated with a first participant of the plurality of participants. The system may collect performance data for each of the plurality of participants within the computer generated multi-realm reality, wherein each of the plurality of participants have performance measurements associated with one or more performance data groupings.

Abstract: Button-suture assemblies which employ a knotted locking mechanism do not maintain tension when securing bone and tissue fragments. A knotless method and device for maintaining tension and providing precise placement of a button-suture assembly during stabilization procedures comprising a locking pin, which mates with a button creating pinch points with increased surface area through which suture can pass from a baseplate through two dedicated openings in the button, one for the first end of the suture and one for the second end of the suture while preventing tension loss during the locking step and therefore allowing for precise placement of the button-suture assembly.

Abstract: Disclosed herein are compounds and ligands, and compositions formed therewith, that modulate seminal endocrine factors that control blood pressure and, thereby, treat hypertension. Also disclosed herein are methods for using the compositions to treat hypertension and disorders associated therewith.

Abstract: The disclosure herein describes platform-level checkpointing for deep learning (DL) jobs. The checkpointing is performed through capturing two kinds of state data: (i) GPU state (device state), and (ii) CPU state (host state). The GPU state includes GPU data (e.g., model parameters, optimizer state, etc.) that is located in the GPU and GPU context (e.g., the default stream in GPU, various handles created by the libraries such as DNN, Blas, etc.). Only a fraction of the GPU memory is copied because the checkpointing is done in a domain-aware manner. The “active” memory contains useful data like model parameters. To be able to capture the useful data, memory management is controlled to identify which parts of the memory are active. Also, to restore the destination GPU to the same context/state, a mechanism is used to capture such state-changing events on an original GPU and replayed on a destination GPU.

Abstract: This disclosure relates to protecting the security of information in content selection and distribution. In one aspect, a method includes receiving, from a client device and by a first computing system of multi-party computation (MPC) systems, a digital component request including first secret shares of data identifying user groups that include a user of the client device as a member. The first computing system transmits a contextual digital component request to a content platform. The first computing system receives, from the content platform, selection data for multiple digital components. The selection data includes first vector data defining a contextual-based vector of values selected based in part on the set of contextual signals. The first computing system obtains, for each digital component, second vector data defining a user group-based vector of values selected based in part on a respective user group corresponding to the digital component.

Abstract: An apparatus calculates a block vector difference (BVD) between a block vector (BV) and a block vector predictor (BVP). The apparatus determines each of a plurality of BV candidates based on a sum of the BVP and a different one of a plurality of BVD candidates. Each of the plurality of BVD candidates comprises: a first and a second component magnitude of the BVD; and a different combination of signs for the first and second component magnitudes. The apparatus determines validities of the plurality of BV candidates based on values of samples, stored in a reference memory, of reference blocks indicated by the plurality of BV candidates. The apparatus signales, in a bitstream, a sign for at least one of the first and second component magnitudes of the BVD based on the validities of the plurality of BV candidates.

Abstract: Provided are methods and systems for managing, distributing, and publishing content. An exemplary method can comprise receiving, in a content management system, an asset for distribution. The method can also comprise generating a metadata object according to a submission standard or requirement of a destination asset distribution platform. The destination asset distribution platform can be external to the content management system and the metadata object can comprise data indicative of or associated with the asset.

Abstract: The technologies described herein are generally directed to validating the integration of new carrier components at base station equipment, e.g., in a fifth generation (5G) network or other next generation networks. For example, a method described herein can include, confirming based on installation data, an installation milestone of an installation of a carrier component at base station equipment, resulting in a confirmed milestone. The method can further include, based on the confirmed milestone, identifying for an evaluation of the installation of the carrier component, a criterion associated with a characteristic of operation of the base station equipment. Further, the method can include based on a result of evaluation of the installation, sending, by the network equipment, a validation signal to the base station equipment indicating that the installation of the carrier component is validated.

Abstract: System and method of generating an executable action item in response to natural language dialogue are disclosed herein. A computing system receives a dialogue message from a remote client device of a customer associated with an organization, the dialogue message comprising an utterance indicative of an implied goal. A natural language processor of the computing system parses the dialogue message to identify one or more components contained in the utterance. The planning module of the computing system identifies the implied goal. The computing system generates a plan within a defined solution space. The computing system generates a verification message to the user to confirm the plan. The computing system transmits the verification message to the remote client device of the customer. The computing system updates an event queue with instructions to execute the action item according to the generated plan upon receiving a confirmation message from the remote client device.

Abstract: Disclosed embodiments provide systems and techniques for mass execution of analytical models across multiple dimensions of client, collateral, deal structure, third party, and other data relevant to predicting optimal decisions in real-time. In some embodiments, disclosed systems and techniques increase decisioning speed through the reduction of computational loads on disclosed decisioning systems. Further disclosed systems and techniques may scale-out analytical modeling computations through, among other technological solutions, advanced execution environments that are asynchronous and non-blocking in nature so as to allow the execution of a plurality of analytical models in parallel and optimizing the results.

Abstract: An image sensor may include an array of image sensor pixels. The array of image sensor pixels may be controlled by row driver circuitry. The row driver circuitry may include row drivers that receive power supply signals from transconductance amplifier circuitry. The transconductance amplifier circuitry may include multiple amplifiers with output ports shorted to one another. Each amplifier may include input transistors, cross-coupled transistors with a low threshold voltage, and additional transistors coupled in series with the cross-coupled transistors and having a moderate or high threshold voltage.

Abstract: A wireless device may receive one or more radio resource control (RRC) messages comprising configuration parameters of a first uplink carrier and a second uplink carrier of an unlicensed cell. Based on switching from the first uplink carrier to the second uplink carrier and a priority class of a listen-before-talk (LBT) procedure for a first transmission of a transport block on the second uplink carrier, the wireless device may determine a contention window to be a minimum contention window of a plurality of contention windows for the LBT procedure. The wireless device may transmit, via the second uplink carrier, the transport block based on the LBT procedure using the contention window.

Abstract: A system for using mobile data to improve weather information is provided. The system includes a weather prediction station configured to receive stationary observation data provided by a plurality of stationary weather stations along with data from a plurality of input weather models and generate unified weather model estimates based on the stationary observation data, the input weather model data, and a processor. The processor is configured to aggregate mobile observation data provided by a plurality of non-stationary sensors and use the aggregated mobile observation data to adjust the weather model estimates.

Abstract: The disclosure provides a botanical processing module, comprising a housing, wherein the housing comprises a plurality of surfaces, wherein there is an internal cavity defined between the plurality of surfaces; a first inlet disposed on a top surface of the housing; a second inlet disposed on a first side surface of the housing, wherein the first side surface is orthogonal to the top surface; a first outlet disposed on the first side surface; a controller, wherein the controller comprises a processor; a memory; and a display, wherein the display is disposed on a second side surface of the housing, wherein the second side surface is orthogonal to both the top surface and the first side surface; a grinder; and a first heating chamber; wherein the grinder and the first heating chamber are disposed within the internal cavity of the housing, wherein the first heating chamber is coupled to the grinder.