Abstract: A method for real-time THz sensing using true time delay (TTD) is implemented by a base station and includes transmitting, by a transceiver that includes TDD elements and phase shifters configured in the transceiver, simultaneous frequency dependent (SFD) beams to scan an environment at a first granularity to detect a spatial cluster target. Each of the SFD beams corresponds to a different phase angle and different frequency. The method includes determining, among the SFD beams, a subset of beams that detected the spatial cluster target. The method includes beam switching, by the transceiver, using time division multiplexing (TDM) and a TDM bandwidth to scan a portion of the environment at phase angles corresponding to the subset of beams and at a second granularity finer than the first granularity. The method includes combining data received from the SFD beams, by multiple threads that concurrently process data received from the SFD beams.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for scalable three-dimensional (3-D) object recognition in a cross reality system. One of the methods includes maintaining object data specifying objects that have been recognized in a scene. A stream of input images of the scene is received, including a stream of color images and a stream of depth images. A color image is provided as input to an object recognition system. A recognition output that identifies a respective object mask for each object in the color image is received. A synchronization system determines a corresponding depth image for the color image. A 3-D bounding box generation system determines a respective 3-D bounding box for each object that has been recognized in the color image. Data specifying one or more 3-D bounding boxes is received as output from the 3-D bounding box generation system.

Abstract: The disclosed technologies are capable of selectively using data structure operations for path query evaluation. One technique involves reading a query that traverses at least two nodes and at least one edge of a graph in a graph database; compiling the query into a set of variables and a set of constraints, where the set of variables and the set of constraints correspond to the two nodes and the one edge of the graph; creating an in-memory data structure that comprises a table; using the set of variables and the set of constraints to determine an operation that is performable using the in-memory data structure; checking for an existence of a condition relating to the in-memory data structure or the operation; skipping the operation if the condition exists or executing the operation if the condition does not exist; and storing a set of intermediate query results in the table.

Abstract: Embodiments described herein provide techniques for in-line compiling of UDFs in other programming languages. These techniques enable faster programming iterations because it allows users to compile directly in the cloud processing system. Moreover, it allows the UDFs to tie into existing libraries. The compiled results are treated as sensitive and handled with appropriate security policies, as with any other user data in the system.

Abstract: Methods and systems are provided for providing directional assistance to guide a user to position a camera for centering a person's face within the camera's field of view. A neural network system is trained to determine the position of the user's face relative to the center of the field of view as captured by an input image. The neural network system is trained using training input images that are generated by cropping different regions of initial training images. Each initial image is used to create a plurality of different training input images, and directional assistance labels used to train the network may be assigned to each training input image based on how the image is cropped. Once trained, the neural network system determines a position of the user's face, and automatically provides a non-visual prompt indicating how to center the face within the field of view.

Abstract: Disclosed are systems, apparatuses, methods, and computer-readable media to train a neural network model implemented into a perception stack in an autonomous vehicle (AV) for detecting objects. A method includes pretraining an uninitialized ML model to yield a first ML model; training the first ML model with a first testing dataset for a first number of iterations based on a first configuration; analyzing the first ML model based on a convergence of the first ML model and a previous iteration of training; generating a report based on the analysis of the first ML; and after generating the report, training the first ML model to yield a second ML model.

Abstract: An electrolyzer system including a hotbox, one or more stacks disposed within the hotbox, a fuel exhaust conduit that receives fuel exhaust output by the stack, a fuel exhaust separator that separates liquid from the fuel exhaust, and a recycling conduit that fluidly connects the fuel exhaust to the fuel inlet conduit.

Abstract: A method for generating a path for wire arc additive manufacturing is provided in this disclosure, which relates to the technical field of additive manufacturing, and includes following steps: generating a model in which a three-dimensional model is established according to angle constraint of the wire arc additive manufacturing; layering the model in which the three-dimensional model is layered along a height direction; selecting discrete points in which a plurality of discrete points are selected according to curve curvature for different layers of the model; obtaining coordinates of the discrete points; determining a printing direction; obtaining coordinates of the discrete points and corresponding printing directions; and generating a control program. The method according to the disclosure is simple, has a wide application range, can satisfy printing of complex shapes, and can serve to well form for structures with maximum printing inclination of 60 degrees, thus improving forming effect of printing.

Abstract: A method includes generating first and second series of segmentation masks for a first and second series of images in a video, respectively. The first series of segmentation masks are generated by using a machine-learning model to (1) generate a first segmentation mask based on a first image in the first series of images and a predetermined fixed segmentation mask, and (2) generate a second segmentation mask based on a second image in the first series of images and the first segmentation mask. The second series of segmentation masks are generated by using the machine-learning model to (1) generate a third segmentation mask based on a third image in the second series of images and the predetermined fixed segmentation mask, and (2) generate a fourth segmentation mask based on a fourth image in the second series of images and the third segmentation mask.

Abstract: Generally discussed herein are systems, devices, and methods that include an organic high density interconnect structure and techniques for making the same. According to an example a method can include forming one or more low density buildup layers on a core, conductive interconnect material of the one or more low density buildup layers electrically and mechanically connected to conductive interconnect material of the core, forming one or more high density buildup layers on an exposed low density buildup layer of the one or more low density buildup layers, conductive interconnect material of the high density buildup layers electrically and mechanically connected to the conductive interconnect material of the one or more low density buildup layers, and forming another low density buildup layer on and around an exposed high density buildup layer of the one or more high density buildup layers.

Abstract: A processor may receive a request with a column-oriented operation. The query request may be related to updates to a row-oriented database based on the query request previously performed on the row-oriented database. The processor may determine the amount of time for replicating the updates to a column-oriented database. If the amount of time for replicating the updates to the column-oriented database is greater than a threshold amount of time, then the processor may create a transient file with the updates. The transient file may be in a format that is recognized by the column-oriented database. The processor may query the column-oriented database based on the query request and the transient file.

Abstract: Techniques are disclosed for using both feature delineation and machine learning to detect cardiac arrhythmia. A computing device receives cardiac electrogram data of a patient sensed by a medical device. The computing device obtains, via feature-based delineation of the cardiac electrogram data, a first classification of arrhythmia in the patient. The computing device applies a machine learning model to the received cardiac electrogram data to obtain a second classification of arrhythmia in the patient. As one example, the computing device uses the first and second classifications to determine whether an episode of arrhythmia has occurred in the patient. As another example, the computing device uses the second classification to verify the first classification of arrhythmia in the patient. The computing device outputs a report indicating that the episode of arrhythmia has occurred and one or more cardiac features that coincide with the episode of arrhythmia.

Abstract: Integrated circuit transistor structures are disclosed that reduce n-type dopant diffusion, such as phosphorous or arsenic, from the source region and the drain region of a germanium n-MOS device into adjacent shallow trench isolation (STI) regions during fabrication. The n-MOS transistor device may include at least 75% germanium by atomic percentage. In an example embodiment, the structure includes an intervening diffusion barrier deposited between the n-MOS transistor and the STI region to provide dopant diffusion reduction. In some embodiments, the diffusion barrier may include silicon dioxide with carbon concentrations between 5 and 50% by atomic percentage. In some embodiments, the diffusion barrier may be deposited using chemical vapor deposition (CVD), atomic layer deposition (ALD), or physical vapor deposition (PVD) techniques to achieve a diffusion barrier thickness in the range of 1 to 5 nanometers.

Abstract: A user equipment (UE) reduces power consumption by selectively disconnecting from a radio access network (RAN) while in a non-standalone (NSA) mode in response to detecting a threshold number of communication failures associated with the RAN. The UE connects to a core network in the NSA mode via both a 4G RAN and a 5G RAN. The UE monitors the connection with the 5G RAN for communication failures, such as Random Access Channel (RACH) failures, radio link failures (RLFs), and the like. In response to the number of detected failures exceeding a threshold, the UE can determine that the quality of the connection with the 5G RAN is providing relatively little benefit and can terminate the connection with the 5G RAN.

Abstract: Systems and methods are provided for detecting a flashing light on one or more traffic signal devices. The method includes capturing a series of images of one or more traffic signal elements in a traffic signal device over a length of time. The method further includes, for each traffic signal element, analyzing the series of images to determine one or more time periods at which the traffic signal element is in an on state or an off state, and analyzing the time periods to determine one or more distinct on states and one or more distinct off states. The method further includes identifying one or more cycles correlating to a distinct on state immediately followed by a distinct off state, or a distinct off state immediately followed by a distinct on state, and, upon identifying a threshold number adjacent cycles, classifying the traffic signal element as a flashing light.

Abstract: Aspects of the present disclosure are directed to orchestration of tasks in tenant clouds. In an embodiment, an orchestrator receives a task-group and a condition, with the task-group specifying multiple tasks. The orchestrator selects a group of resources satisfying the condition, with the group of resources being of a tenant cloud spanning multiple cloud infrastructures. The orchestrator invokes the task-group on the group of resources to cause each of the multiple tasks to be executed on each of the group of resources.

Abstract: Disclosed herein are systems and methods to perform hybridized transmission switching of an electric power system to avoid exceeding line ratings and minimize load shedding.

Abstract: A method includes: identifying, by a runtime instrumentation agent of a web server, a plurality of attack surfaces of a web application executed on the web server; generating, by the runtime instrumentation agent, a plurality of hash values, where each hash value is generated based on one of the plurality of attack surfaces; and transmitting, by the runtime instrumentation agent, the plurality of hash values to an attack server external to the web server, where the attack server is to determine whether to scan each attack surface based on the plurality of hash values.

Abstract: Proactive caching, at a client device (e.g., a feature phone), of transient assistant action suggestions for selective rendering by an assistant client application of the client device. A transient assistant action suggestion, when rendered via an assistant client application and selected, causes the assistant client application to initiate performance of a corresponding assistant action. In various implementations, a prefetched transient action suggestion can be a time-constrained suggestion that includes at least associated rendering restriction metadata that defines one or more temporal windows to which rendering of the time-constrained suggestion is restricted. Proactive cache refresh rate metadata can also be associated with transient action suggestion(s) and defines a duration during which the assistant client application is to refrain from interfacing with a remote system to prefetch updated transient assistant action suggestions.

Abstract: Techniques are disclosed for using feature delineation to reduce the impact of machine learning cardiac arrhythmia detection on power consumption of medical devices. In one example, a medical device performs feature-based delineation of cardiac electrogram data sensed from a patient to obtain cardiac features indicative of an episode of arrhythmia in the patient. The medical device determines whether the cardiac features satisfy threshold criteria for application of a machine learning model for verifying the feature-based delineation of the cardiac electrogram data. In response to determining that the cardiac features satisfy the threshold criteria, the medical device applies the machine learning model to the sensed cardiac electrogram data to verify that the episode of arrhythmia has occurred or determine a classification of the episode of arrhythmia.