Abstract: A test system includes a computer system including a processor and a memory; and an electronic interface in communications with the computer and a television device under test, the electronic interface including a test interface board configured to permit the television device under test to be put into a debug mode by the computer system.

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: In various embodiments, a method for estimating a room layout includes receiving candidate wall distance estimates for walls in an acoustic environment and confidence scores; receiving device location data for devices; determining, based on the device location data, a location of each wall relative to each audio processing device; generating a room layout matrix that includes the candidate wall distance estimates, wherein each row of the room layout matrix is associated with a respective device and each column is associated with a respective wall and each of the candidate wall distance estimates is associated with the respective walls based on the confidence scores; determining, based on a highest confidence score in each column of the room layout matrix and the location data, a set of wall distance estimates from an ordinal device; and determining, based on the set of wall distance estimates, room layout estimates of the acoustic environment.

Abstract: In various embodiments, a computer-implemented method comprises emitting a reference audio signal using one or more speakers of an audio processing device located in an acoustic environment, receiving, by each microphone in a microphone pair of the audio processing device, a response of the acoustic environment to the reference audio signal, determining an impulse response of the acoustic environment based on the reference audio signal and the received response of the acoustic environment, determining, based on the impulse response, a plurality of candidate wall distance estimates and a plurality of confidence scores, and determining based on the plurality of candidate wall distance estimates and confidence scores, including candidate wall distance estimates and confidence scores from one or more other audio processing devices, a set of wall distance estimates, each wall distance estimate in the set of wall distance estimates being associated with a different wall in a plurality of walls.

Abstract: An electronic device and a method for digitization of human motion data for semantic search is provided. The electronic device receives a sequence of input images depicting a physical performance associated with a body movement discipline and extracts a localized instance of a performer associated with the physical performance from an image frame of the sequence of image frames. The electronic device determines a plurality of performance attributes associated with the performer based on application of an attribute recognition network on the localized instance. The plurality of performance attributes comprises an emotion attribute, a posture attribute, and a hand gesture attribute. The electronic device searches a performance database based on the plurality of performance attributes to generate a search result. The search result specifies whether the localized instance represents an unregistered pose or a registered pose of the body movement discipline.

Abstract: A secure data recorder provides targeted collection and storage of working data from any subsystem of a computing device. The data recorder gathers and stores device working data based on stored configuration data. The configuration data indicates memory and storage locations on the device from which to gather working data and storage locations at which the data recorder stores the gathered working data. The data recorder operates in a secure execution environment during all of the pre-boot stage of the computing device. The data recorder further allows a user to update the Basic Input/Output System (BIOS) of the computing device based on a firmware image that may be received via the network. The data recorder also facilitates gathering of working data, over time, that may reveal a malfunction of particular hardware configurations and particular software configurations.

Abstract: A test system includes a computer system including a processor and a memory; and an electronic interface in communications with the computer and a television device under test, the electronic interface including a test interface board configured to permit the television device under test to be put into a debug mode by the computer system.

Abstract: A method is disclosed for exporting a black-box model of an inverter-based resource or plant from a first software platform for use by a second software platform. The method comprises simulating, using the first software platform, instantaneous time-domain responses of the inverter-based resource or plant to respective conditions defined by a script, according to the black-box model. The method may further comprise generating training data from the instantaneous time-domain responses and the respective conditions. The method may further comprise, with the training data, training a machine learning model to model the inverter-based resource or plant, wherein the trained machine learning model is transparent as to its inner workings. The method may further comprise generating software code that represents the trained machine learning model in terms of software code usable for defining a custom model of the inverter-based resource or plant in the second software platform.

Abstract: Disclosed embodiments include techniques for determining spatial impulse response via acoustic scrambling. These techniques include a computer-implemented method for generating a frequency sweep signal, the method comprising generating a frequency sweep signal having a monotonically increasing frequency, partitioning the frequency sweep signal into N input segments, each of the N input segments representing a different frequency range, generating an encoding key having a sequence of N non-consecutive numbers, wherein each number in the sequence appears once, generating an output signal by selecting each of the N input segments in an order based on the sequence of N non-consecutive numbers in the encoding key, and causing a speaker to produce audio tones in an audio space based on the output signal.

Abstract: The disclosure relates to methods and system for handling immersive services with feature tags in an 3GPP IP multimedia subsystem (IMS) and 3GPP mission critical services (MCX). A user equipment uses the media tags, feature tags with unique IMS communication service identifier (ICSI) and IMS application reference identifier (IARI) to communicate with the IMS network for determining the immersive services in the communication. Accordingly, a method includes: performing, a registration with a network for handling immersive call services based on feature tags included in a session initiation protocol (SIP) REGISTER request; determining whether the network provides a provision for the handling of immersive call services based on an SIP REGISTER response; and sending an SIP INVITE request including feature tags to a second client device via the network based on determining that the network provides the provision of handling immersive media streams.

Abstract: A system and method for classifying network traffic including: providing a framework of models configured to classify network traffic; determining, via a packet processing engine, whether the network traffic flow is encrypted; if the network traffic flow is encrypted: extracting, via the packet processing engine, encrypted flow attributes associated with a network traffic flow; deriving further flow attributes based on the extracted encrypted flow attributes; determining at least one model of the framework of models based on the derived flow attributes and extracted encrypted flow attributes; and classifying, via a traffic classification engine, the traffic flow as a category of traffic based on the at least one model; and if the network traffic flow is not encrypted: extracting non-encrypted flow attributes associated with a network traffic flow; classifying the traffic flow as a category of traffic based on the non-encrypted flow attributes.

Abstract: Embodiments improve data quality using artificial intelligence. Incoming data that includes a plurality of rows of data and a trained neural network that is configured to predict a data category for the incoming data can be received, where the neural network has been trained with training data including training features, and the training data includes labeled data categories. The incoming data can be processed, where the processing extracts features about the plurality of rows of data to generate metadata profiles that represent the incoming data. Using the trained neural network, a data category for the incoming data can be predicted, where the prediction is based on the generated metadata profiles.

Abstract: Systems and methods for optimal split federated learning (O-SFL) in a wireless network, including: receiving, by a federal device in the wireless network, local split points associated with a deep neural network (DNN) model over a time period from at least one client device of a plurality of client devices, wherein the plurality of client devices are connected to an edge device for training the DNN model using split federated learning (SFL); determining, by the federal device, an average of the local split points; determining, by the federal device, a global split point for partitioning the DNN model between the at least one client device and the edge device based on the average of the local split points; and applying, by the federal device, the determined global split point to train the DNN model.

Abstract: An electronic device and a method for implementation for frame-anomaly based video shot segmentation using self-supervised machine learning (ML) model is disclosed. The electronic device receives video data including a set of video frames and creates a synthetic shot dataset including a set of synthetic shots. The electronic device pre-trains an ML model and selects the training data including a first subset of video frames corresponding to a first synthetic shot. The electronic device fine-tunes the pre-trained ML model and selects a test video frame. The electronic device applies the fine-tuned ML model on the test video frame to determine whether the test video frame corresponds to an anomaly. The electronic device labels the first subset of video frames as a single shot. The set of video frames is segmented into a set of shots. The electronic device controls a rendering of the set of shots on a display device.

Abstract: Methods, systems, and computer-readable media are disclosed herein that utilizes historical changes made to files and methods in computer programming code to predict related files and methods that may be affected by current and/or future changes made to other files and methods. In aspects, when computer code for a particular method is going to be edited, other methods are identified that were changed in previous editing sessions that also included changes to the particular method. Using scoring techniques for the other methods, a recommendation is provided that details the relative strength of whether the other methods are predicted to be affected by any changes made to the computer code for the particular method that is edited.

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: The boosting of the weights related to imperfect matches of electronic records from disparate sources is discussed. The imperfect matches may be in primary data (such as a code) and/or in supplemental data between two or more records that correspond to the same person (such as a patient). The imperfect matches are analyzed to determine whether they are sufficient to warrant de-duplication of those imperfect matches in a final combined record for the person. The boosting of the weights may be based upon any of numerous factors, such as various distance measures between the supplemental information as a measure of how different the supplemental information is between the respective records.

Abstract: Systems and methods for amplifying a signal is described. A circuit may convert an input radio frequency (RF) signal into a first RF signal with power level matching a power capacity of a first transistor of a first size in a carrier amplifier stage, a second RF signal with power level matching a power capacity of a second transistor of the first size in a peaking amplifier stage, and a third RF signal with third power level matching a power capacity of a third transistor of a second size in another peaking amplifier stage. The circuit may amplify the first, second, and third RF signals to generate first, second, and third amplified RF signals, respectively. The circuit may combine the first, second, and third amplified RF signals, into an output RF signal that is an amplified version of the input RF signal.

Abstract: The embodiments described herein provide means for replicating external stages between deployments of e.g., a cloud data lake using a modified storage integration. The modified storage integration may be defined with a set of storage locations, wherein the storage integration comprises a base URL for each of the set of storage locations and wherein each storage location identifies a remote deployment where a cloud platform is hosted and a geographic region of the remote deployment. An external stage object may be bound to the storage integration, wherein the external stage object facilitates a data loading operation that is currently in progress on the first storage location. In response to detecting an outage at the first storage location, the data loading operation that was in progress on the first storage location may be continued at the second storage location using the storage integration.

Abstract: An audio system is provided. The audio system includes a plurality of loudspeakers and a mobile device. The plurality of loudspeakers is capable of being positioned in a listening environment and being arranged to transmit an audio signal in the listening environment. Each loudspeaker is programmed to determine angle information of the audio signal as received at the loudspeaker relative to other loudspeakers of the plurality of loudspeakers and to transmit a first signal indicative of the angle information. The mobile device is programmed to receive the first signal from each of the loudspeakers and to determine a location for each loudspeaker in the listening environment based at least on the angle information.