Abstract: Presented herein are embodiments of a training deep learning models. In one or more embodiments, a compact deep learning model comprises fewer layers, which require fewer floating-point operations (FLOPs). Presented herein are also embodiments of a new learning rate function, which can adaptively change the learning rate between two linear functions. In one or more embodiments, combinations of half-precision floating point format training together with larger batch size in the training process may also be employed to aid the training process.

Abstract: Methods, apparatuses, and systems for wireless communications are described. Uplink signals may be time aligned based on different synchronization signals as a timing reference for different carrier groups. A base station may send, to a wireless device, a time alignment command comprising: an indication of a carrier group index, and a time adjustment value. Based on the time alignment command, the wireless device may send second uplink signals.

Abstract: Computing devices for automatic placement and arrangement of objects in computer-based 3D environments are disclosed herein. In one embodiment, a computing device is configured to provide, on a display, a user interface containing a work area having a template of a 3D environment and a gallery containing models of two-dimensional (2D) or 3D content items. The computing device can then detect, via the user interface, a user input selecting one of the models from the gallery to be inserted as an object into the template of the 3D environment. In response to detecting the user input, the computing device can render and surface on the display, a graphical representation of the 2D or 3D content item corresponding to the selected model at a location along a circular arc spaced apart from the default viewer position of a viewer of the 3D environment by a preset radial distance.

Abstract: A system for classifying a target image with segments having attributes is provided. The system generates a graph for the target image that includes vertices representing segments of the image and edges representing relationships between the connected vertices. For each vertex, the system generates a subgraph that includes the vertex as a home vertex and neighboring vertices representing segments of the target image within a neighborhood of the segment represented by the home vertex. The system applies an autoencoder to each subgraph to generate latent variables to represent the subgraph. The system applies a machine learning algorithm to a feature vector comprising a universal image representation of the target image that is derived from the generated latent variables of the subgraphs to generate a classification for the target image.

Abstract: A pedicle screw fastening system that consists of an anchoring bone screw having a first diameter with threads on one end for use in anchoring the screw to the bone and a spherical ball connector member having a second diameter on the other end, with the second diameter smaller than the first diameter; the spherical ball connector member operating as a pivot point about which a connecting assembly moves in a polyaxial fashion. The connecting assembly has a lower connector member with an attachment ring to secure the spherical ball connector member, allowing the use of the small diameter spherical ball connector member with a large shank anchoring bone screw. A biased retainer member maintains a positive tension between the connecting assembly and the anchored screw. The system allows for large shank screws with small connecting assemblies.

Abstract: Described herein are adhesive support devices, which may include adhesive medical devices. In particular, three-dimensional adhesive medical devices from a single layer of material, typically by stamping or pressing (including cold pressing) a sheet of the material to form rigid and complex 3D shapes capable of use for a variety of medical and non-medical purposes. These adhesive support devices typically also include an adhesive material allowing the device to be attached to a subject's skin and/or mounted on a surface (e.g., wall, etc.). In particular, described herein are methods of manufacturing, assembling, and using adhesive support devices.

Abstract: A harmonic drive assembly and fluid-powered linear motors with both axial pistons are rotary piston arrangements incorporating the harmonic drive assembly are disclosed. The motors may be used in downhole drilling applications, but the drive assembly and/or motors may be used in other applications. The assembly, motors and methods use advanced harmonic drives, advanced helical drives, and combinations thereof with 1) motors with axial pistons and reciprocating linear rings to convert reciprocative axial motion to continuous rotary motion, and 2) motors with rotary pistons and reciprocating linear rings to rectify reciprocative rotary motion to continual rotary motion to improve performance over prior configurations.

Abstract: Techniques are provided for compression of log data using field types. An exemplary method comprises: obtaining at least one log message, wherein the at least one log message comprises a message template and at least one message variable, wherein the message template comprises at least one field corresponding to the at least one message variable; obtaining a compression index that maps a plurality of message templates to a corresponding message signature; identifying a predefined field type of the at least one field; selecting a compression technique to apply to the at least one message variable based on the predefined field type to obtain a compressed message variable; and writing the compressed message variable and a message signature corresponding to the message template of the at least one log message to a log file. The at least one log message can be a historical log message or part of a real-time stream of log messages.

Abstract: Disclosed are preforms which incorporate improvements in the region of the neck and upper segment of the body to allow the production of lightweight containers, such as bottles suitable for containing water or other beverages. In accordance with certain embodiments, the improvements include a thinner neck finish area than conventional bottles, where the thinner area is extended into the upper segment of the body portion below the support ring. Reducing the thickness in these areas of the bottle allows for less resin to be used in forming the preform and bottle.

Abstract: A botanical extract that exhibits anti-inflammatory activity, wherein the botanical extract is at least an extract from the leaf of Vaccinium macrocarpon.

Abstract: A multi-sensor auscultation device is disclosed that can be applied to or worn by a user to monitor multiple physiological systems of the user. The multi-sensor auscultation device can make use of two or more acoustic sensors, such as accelerometer contact microphones (ACMs), to collect acoustic data from multiple locations on the user's body. The multi-sensor auscultation device can include electrodes for detecting cardiac electrical activity and/or assessing the user's bioimpedance. The multi-sensor auscultation device can provide useful data associated with the user's cardiovascular system, respiratory system, and/or electrical characteristics. The multi-sensor auscultation device can be in the form of a reusable electronics module couplable to a disposable patch adhesive.

Abstract: A method of partitioning a video coding block for JVET, comprising representing a JVET coding tree unit as a root node in a quadtree plus binary tree (QTBT) structure that can have a quadtree branching from the root node and binary trees branching from each of the quadtree's leaf nodes using asymmetric binary partitioning to split a coding unit represented by a quadtree leaf node into two child coding units of unequal size, representing the two child coding units as leaf nodes in a binary tree branching from the quadtree leaf node and coding the child coding units represented by leaf nodes of the binary tree with JVET, wherein further partitioning of child coding units split from quadtree leaf nodes via asymmetric binary partitioning is disallowed.

Abstract: A lithium ion battery including an electrolyte and a lithium titanate negative electrode is provided. The lithium titanate negative electrode includes structures of a lithium titanate core and a conformal layer surrounding each lithium titanate core. The conformal layer either includes titanium oxide with substantially no lithium or has a concentration of lithium ranging from a lower concentration at a surface portion of the layer to a higher concentration at an interior portion of the layer adjacent to the lithium titanate core. A method of preparing the lithium titanate structures and a method of preparing an electrode for a lithium ion battery, wherein the electrode includes lithium titanate structures, are also provided.

Abstract: Techniques, methods, systems, and other mechanisms for measuring the excursion of a speaker while being actively driven. Measuring excursion can involve attaching a flexible printed coil (FPC), including a sense coil, to the speaker, and monitoring an induced current as produced though the sense coil, and further detecting that violation of an excursion limit for the speaker may likely occur.

Abstract: A system and method for monitoring the dynamic load moment and stabilization of a lifting vehicle is provided. The system is adapted for monitoring the stability of the vehicle on a real-time basis, and displaying the real-time dynamic load moment of the vehicle/cargo combination on a continuous scale on a user interface. The system may also include an object detection device oriented for determining the presence or absence of a floor surface within a specified distance from the vehicle in order to detect drop-offs in the floor surface surrounding the vehicle and alert the operator of same. A method for calibrating the system is also provided. The calibration method can include a series of lift and acceleration steps for collecting static and dynamic data in order to determine one or more coefficients associated with the vehicle.

Abstract: A resin composition and method for installing a pipe liner that allows the liner to be fully wet out with a resin and activator and stored for a period of up to six months prior to installation and curing. A method of lining a pipe with a delayed curing resin composition is also provided that includes fully wetting out a liner with a blended two part epoxy composition such that the liner can be transported in a wet out fashion, placed in a pipe to be lined and repositioned as needed without concern for the resin composition to begin curing.

Abstract: A device (e.g., an article of athletic gear) comprising a post-molded expandable component, which is a part of the device that is configured to be expanded or has been expanded after being molded. This may allow the post-molded expandable component to have enhanced characteristics (e.g., be more shock-absorbent, lighter, etc.), to be cost-effectively manufactured (e.g., by using less material and/or making it in various sizes), and/or to be customized for a user (e.g., by custom-fitting it to the user).

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for machine learning models for adjusting communication parameters. In some implementations, data for each terminal in a set of multiple satellite terminals is obtained. A machine learning model is trained based on the obtained data. The model can be trained to receive an indication of a geographic location and predict a satellite beam capable of providing at least a minimum level of efficiency for communication at the geographic location. After training the machine learning model, an indication of a predicted satellite beam for a particular location is generated for a particular geographic location. A determination is then made whether to change the current satellite beam for a terminal at the particular geographic location based on the predicted satellite beam.

Abstract: Methods for obtaining a sample from a subject include providing a sample collection room within a retail store; and obtaining a sample from a subject at a retail location. Sample collection rooms may house a sample analysis device or system. Samples may be small, e.g., a finger-stick capillary blood sample.

Abstract: An endovascular delivery device is disclosed. The device comprises a handle assembly at a distal end thereof and a top-cap assembly at a proximal end thereof. The top-cap assembly has an open configuration and a closed configuration and comprises: a capsule comprising an inner portion and an outer tubular portion and defining a cavity within the outer tubular portion, the cavity receiving the proximal portion of the endograft in the open configuration; and a retriever assembly having a body portion arranged and constructed to fit within the cavity of the capsule in the closed configuration. The device further comprises: a guide wire catheter extending through the handle assembly, the guide wire catheter being affixed at a proximal end thereof to the top-cap assembly; and an endograft receiving portion extending distally with respect to the top-cap assembly. In the closed configuration, the cavity is closed by the body portion of the retriever assembly.