Abstract: The subject disclosure relates to systems for managing the deployment and updating of incremental machine learning models across multiple geographic sovereignties. In some aspects, systems of the subject technology are configured to perform operations including: receiving a first machine learning model via a first coordination agent, the first machine learning model based on a first training data set corresponding with a first sovereign region, sending the first machine learning model to a second coordination agent in a second sovereign region, wherein the second sovereign region is different from the first sovereign region, and receiving a second machine learning model from the second coordination agent, wherein the second machine learning model is based on updates to the first machine learning model using a second training data set corresponding with the second sovereign region. Methods and machine-readable media are also provided.

Abstract: A valve prosthetic implant for treating a vein or other blood vessel includes a tubular, expandable anchoring frame extending from a proximal end to a distal end of the implant, a valve seat formed at or near the middle of the anchoring frame, an expandable ball disposed within the lumen of the anchoring frame, and a ball retention tether attached to the expandable ball and to the valve seat and/or the anchoring frame. The anchoring frame may include a cylindrical proximal portion at the proximal end, a cylindrical distal portion at the distal end, an inwardly angled inlet portion between the cylindrical proximal portion and a middle of the anchoring frame, and an inwardly angled outlet portion between the cylindrical distal portion and the middle of the anchoring frame.

Abstract: A vehicle routing system instructs a first vehicle having a human driver to execute a first route for delivering a first payload from a first pick-up location to a first destination. Instructions provided to the first vehicle include an instruction to drop-off the human driver at a driver transfer location along the first route and to continue along the first route without the human driver. The system further instructs a second vehicle that is operating autonomously to execute a second route that also includes the driver transfer location. The instructions to the second vehicle include an instruction to pick-up the human driver from the first vehicle at the driver transfer location and to continue along the second route with the human driver. The route segments may be labeled as suitable or unsuitable for autonomous driving to identify where to locate the driver transfer locations.

Abstract: Chat room content classification, in an online communication environment, where higher applicable chat rooms are prioritized for a user, is provided. First, an chat room service receives chat room content for at least a first chat room and a second chat room. A chat room analyzer can then analyze a characteristic(s) associated with the first chat room and/or the second chat room. Based on the characteristic, the chat room determines that the first chat room is more applicable to the user. Then, a user interface may be presented to the user where the first chat room is prioritized (or ranked) over the second chat room.

Abstract: Systems, methods, and devices are disclosed for detecting an active speaker in a two-way conference. Real time audio in one or more sub band domains are analyzed according to an echo cancellor model. Based on the analyzed real time audio, one or more audio metrics are determined from output from an acoustic echo cancellation linear filter. The one or more audio metrics are weighted based on a priority, and a speaker status is determined based on the weighted one or more audio metrics being analyzed according to an active speaker detection model. For an active speaker status, one or more residual echo or noise is removed from the real time audio based on the one or more audio metrics.

Abstract: This disclosure relates to solutions for eliminating undesired audio artifacts, such as background noises, on an audio channel. A process for implementing the technology can include receiving a set of audio segments, analyzing the segments using a first ML model to identify a first probability of unwanted background noises in the segments, and if the first probability exceeds a threshold, analyzing the segments using a second ML model to determine a second probability that the one or more background features exist in the segments. In some aspects, the process can include attenuating audio artifacts in the segments, if the second probability exceeds a second threshold. In some implementations, dynamic time stretching and shrinking can be applied to the noise attenuation. Systems and machine-readable media are also provided.

Abstract: In an example implementation according to aspects of the present disclosure, a method may include determining how many computing devices are in a stackable configuration and selecting an algorithm for charging the computing devices while in the stackable configuration. The method may include, based on the number of computing devices in the stackable configuration, charging the computing devices according to the selected algorithm.

Abstract: The invention aligns a digital face model from a 3D face scanner with a 3D jaw scan from an intraoral scan produced by an intraoral scanner without using external markers. The alignment proceeds in two steps. In the first step, the teeth part of a subject whose teeth are clenched, referred to as a clenched-teeth face scan, is aligned with an intraoral jaw scan to obtain a first transformation matrix. In the second step, a face model of a subject with a normal facial expression is aligned with the clenched-teeth face model of the same subject to obtain a second transformation matrix. A graphical user interface is provided that enables a user to manually align the 3D jaw scan with a 2D image of the subject's teeth to determine the first transformation.

Abstract: The invention aligns a digital face model from a 3D face scanner with a 3D teeth model from an intraoral scan produced by an intraoral scanner without using external markers. The alignment proceeds in two steps. In the first step, the teeth part of a subject whose teeth are clenched, referred to as a clenched-teeth face scan, is aligned with the teeth model from an intraoral jaw scan to obtain a first transformation matrix. In the second step, a face model of a subject with a normal facial expression is aligned with the clenched-teeth face model of the same subject using an ICP algorithm.

Abstract: A vibration reduction (VR) human support has exactly two paths to ground in each degree of VR: one provided by respective displacement actuators (DAs) for active VR, the other by one or more elastomeric damping bodies (EDBs). These paths extend from a frame for carrying a support structure for a live human, to a grounding for fixing the VR system in a vibrating environment. Each EDB is composed of a material having a dynamic Young's Modulus of 0.1-2.5 MPa, and a resilience test rebound height less than 40, and is positioned between the grounding and frame to provide elastic restorative forces and damping in each of the respective directions. The use of EDBs simplifies construction, and allows for a more compact arrangement, without reducing VR efficiency.

Abstract: Systems, methods, and devices are disclosed for detecting an active speaker in a two-way conference. Real time audio in one or more sub band domains are analyzed according to an echo cancellor model. Based on the analyzed real time audio, one or more audio metrics are determined from output from an acoustic echo cancellation linear filter. The one or more audio metrics are weighted based on a priority, and a speaker status is determined based on the weighted one or more audio metrics being analyzed according to an active speaker detection model. For an active speaker status, one or more residual echo or noise is removed from the real time audio based on the one or more audio metrics.

Abstract: Systems and methods are provided for to estimate the pose of a human subject's head from a sequence of images received from a single depth camera by processing the images to generate a continuous estimate of the head pose in a 3-dimensional (3D) space, and to generate a 3D head model for display and further use. The subject is provided instructions to rotate their head in a first direction until a threshold angle of rotation is reached and then are provided instructions to rotate their head in a second direction. The depth camera provides a sequence of captured images which are processed to extract head meshes. After capture is complete the head meshes are merged to generate a 3D model of the subject's head.

Abstract: An apparatus, system and method are provided for a wireless electrical stimulation. The apparatus generally includes at least two electrical stimulation units. Each electrical stimulation unit includes electrodes connected to the unit. The apparatus also includes a receiver configured for receiving the one or more control signals from a remote controller for remotely, wirelessly controlling each of the electrical stimulation units to selectively apply a time-varying electric potential to the electrodes to provide an electrical stimulation to tissue in electrical contact with the electrodes. The apparatus generally applies to a foot of a human body.

Abstract: A system and method for malware assessment of an unknown application file are provided. The system and method are particularly applicable to malware assessment for Android® operating system based applications. The malware assessment increases the confidence in an assessment that an application is benign or malware.

Abstract: An apparatus, system and method are provided for a wireless electrical stimulation. The apparatus generally includes at least two electrical stimulation units. Each electrical stimulation unit includes electrodes connected to the unit. The apparatus also includes a receiver configured for receiving the one or more control signals from a remote controller for remotely, wirelessly controlling each of the electrical stimulation units to selectively apply a time-varying electric potential to the electrodes to provide an electrical stimulation to tissue in electrical contact with the electrodes. The apparatus generally includes a heating device.