Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a policy neural network having a plurality of policy parameters and used to select actions to be performed by an agent to control the agent to perform a particular task while interacting with one or more other agents in an environment. In one aspect, the method includes: maintaining data specifying a pool of candidate action selection policies; maintaining data specifying respective matchmaking policy; and training the policy neural network using a reinforcement learning technique to update the policy parameters. The policy parameters define policies to be used in controlling the agent to perform the particular task.

Abstract: Apparatuses, methods, and systems are disclosed for NSSAI configuration. One method includes determining a configured network slice selection assistance information for a public land mobile network for a remote unit in response to a trigger from an access and mobility management function. The access and mobility management function, a network slice selection function, or a combination thereof determines the configured network slice selection assistance information. The method includes providing the configured network slice selection assistance information to the remote unit via the access and mobility management function.

Abstract: Neural wirings may be discovered concurrently with training a neural network. Respective weights may be assigned to each edge connecting nodes of a neural graph, wherein the neural graph represents a neural network. A subset of edges may be designated based on the respective weights and data is passed through the neural graph in a forward training pass using the designated subset of edges. A loss function may be determined based on the results of the forward training pass and parameters of the neural network and the respective weights assigned to each edge may be updated in a backwards training pass based on the loss function. The steps of designating the subset of edges, passing data through the neural graph, determining the loss function, and updating parameters of the neural network and the respective weights may be repeated to train the neural network.

Abstract: An image processing method includes the following steps: dividing an object block into a two-dimensional image; identifying at least one view hotspot in a viewing field corresponding to pupil gaze direction; receiving the view hotspot and an indicator signal; wherein the indicator signal is used to remark the object block; and generating a mask block that corresponds to the object block according to the view hotspot; wherein the indicator signal determines the label of the mask block.

Abstract: Embodiments of the present disclosure include systems and methods for training neural networks. In one embodiment, neural network may receive input data and produce output results in response to the input data and weights of the neural network. An error is determined at an output of the neural network based on the output results. The error is propagated in a reverse direction through the neural network from the output and one or more intermediate outputs to adjust the weights.

Abstract: The technology disclosed corrects inter-cluster intensity profile variation for improved base calling on a cluster-by-cluster basis. The technology disclosed accesses current intensity data and historic intensity data of a target cluster, where the current intensity data is for a current sequencing cycle and the historic intensity data is for one or more preceding sequencing cycles. A first accumulated intensity correction parameter is determined by accumulating distribution intensities measured for the target cluster at the current and preceding sequencing cycles. A second accumulated intensity correction parameter is determined by accumulating intensity errors measured for the target cluster at the current and preceding sequencing cycles. Based on the first and second accumulated intensity correction parameters, next intensity data for a next sequencing cycle is corrected to generate corrected next intensity data, which is used to base call the target cluster at the next sequencing cycle.

Abstract: A system and method for providing unsupervised domain adaption for spatio-temporal action localization that includes receiving video data associated with a source domain and a target domain that are associated with a surrounding environment of a vehicle. The system and method also include analyzing the video data associated with the source domain and the target domain and determining a key frame of the source domain and a key frame of the target domain. The system and method additionally include completing an action localization model to model a temporal context of actions occurring within the key frame of the source domain and the key frame of the target domain and completing an action adaption model to localize individuals and their actions and to classify the actions based on the video data. The system and method further include combining losses to complete spatio-temporal action localization of individuals and actions.

Abstract: Some methods enable a computer to receive location information from a device and determine a coverage region for the device. The device's coverage region is the area within range of a wireless signal coming from the device. The computer can determine that part of the device's coverage region covers part of a specified region and instruct the device to provide wireless access to a network associated with the first region.

Abstract: A system and method in an intelligent network are provided for analyzing an event external to the intelligent network. A set of intelligent network subscribers is selected in a selected geographic area. Next, signal message data is obtained from a signaling network in the intelligent network, wherein the signal message data indicates status of subscribers in the set of intelligent network subscribers. External data related to the selected geographic area is obtained, wherein the external data is external to the intelligent network. The signal message data and the external data are then fused to produce synthetic data. In response to the synthetic data, an alert of an external event is produced. The alert can be graphically represented on a display.

Abstract: The present disclosure relates to utilizing a neural network having a two-stream encoder architecture to accurately generate composite digital images that realistically portray a foreground object from one digital image against a scene from another digital image. For example, the disclosed systems can utilize a foreground encoder of the neural network to identify features from a foreground image and further utilize a background encoder to identify features from a background image. The disclosed systems can then utilize a decoder to fuse the features together and generate a composite digital image. The disclosed systems can train the neural network utilizing an easy-to-hard data augmentation scheme implemented via self-teaching. The disclosed systems can further incorporate the neural network within an end-to-end framework for automation of the image composition process.

Abstract: Certain embodiments of the disclosure may include devices, systems, and methods for photographing and dimensioning objects in a photo booth. The devices may include multiple sets of components, and the multiple sets of components may be mounted in different locations of the booth as desired and as needed, for example, depending on the object to be photographed and dimensioned. The photo booth components and functions can be operated and managed via central processing and via user interface and can be in communication with remote computers.

Abstract: A control system includes a communications unit, an acquisition unit, and an output unit. The communications unit communicates with a transmitter that transmits a wireless signal. The acquisition unit acquires location information about a location of the transmitter based on the wireless signal that has been transmitted from the transmitter and received by the communications unit. The output unit has an instruction signal including adjustment information transmitted, in accordance with the location information acquired by the acquisition unit, from the communications unit to the transmitter. The adjustment information is used to vary at least one of a time interval at which the transmitter transmits the wireless signal or transmission power with which the transmitter transmits the wireless signal.

Abstract: Systems and methods for object recognition with limited input are disclosed herein. An example method includes updating a neural network trained to perform object recognition on a first rendition of an object, so that the neural network performs object recognition on a second rendition of the object, using a limited set of input images. The method includes receiving a limited set of model images of the second rendition of the object, accessing a corresponding image mapping, and generating a large number of training images from the limited set, where image mappings include geometric, illumination, and/or obscuration transformations. The neural network is then trained, from this initial small set, to classify the second rendition of the object.

Abstract: The invention provides an automatic inspection process for detecting visible defects on a manufactured item. The process includes a set up mode in which images of same-type defect free items, but not images of same-type defected items, are obtained, and an inspection mode in which images of both same-type defect free items and same-type defected items, are obtained and defects are detected. Images of the same-type defect free items are analyzed and based on the analysis the process switches to the inspection mode.

Abstract: A communication apparatus receives a signal that is transmitted from a partner apparatus using a first communication scheme, and establishes, on the basis that the signal received includes information indicating that parameter exchange processing is to be executed, a connection with the partner apparatus using a second communication scheme, wherein the second communication scheme has a higher communication rate and consumes more power than the first communication scheme. In addition, the communication apparatus exchanges, by using the second communication scheme, information to be used in communication for the parameter exchange processing with the partner apparatus, and executes the parameter exchange processing using the information by using the second communication unit scheme.

Abstract: Implementations are described herein for automatically generating synthetic training images that are usable as training data for training machine learning models to detect, segment, and/or classify various types of plants in digital images. In various implementations, a digital image may be obtained that captures an area. The digital image may depict the area under a lighting condition that existed in the area when a camera captured the digital image. Based at least in part on an agricultural history of the area, a plurality of three-dimensional synthetic plants may be generated. The synthetic training image may then be generated to depict the plurality of three-dimensional synthetic plants in the area. In some implementations, the generating may include graphically incorporating the plurality of three-dimensional synthetic plants with the digital image based on the lighting condition.

Abstract: An electronic device is provided. The electronic device includes a first communication circuit, a second communication circuit, a processor configured to be electrically connected with the first communication circuit and the second communication circuit, and a memory configured to be electrically connected with the processor. The memory includes instructions, when executed by the processor, cause the processor to obtain location information of the electronic device, transmit a first message for requesting to change a state of the electronic device to a network, receive a first response message to the transmitted first message from the network, transmit a second message for requesting a parameter for an operation cycle of the second communication circuit to the network, receive a second response message to the second message from the network, and change the operation cycle of the second communication circuit to a value corresponding to a current state of the electronic device.

Abstract: A method for computing a dominant class of a scene includes: receiving an input image of a scene; generating a segmentation map of the input image, the segmentation map being labeled with a plurality of corresponding classes of a plurality of classes; computing a plurality of area ratios based on the segmentation map, each of the area ratios corresponding to a different class of the plurality of classes of the segmentation map; and outputting a detected dominant class of the scene based on a plurality of ranked labels based on the area ratios.

Abstract: A user device cradle can include a receiver configured to removably retain a wireless user device. One or more Radio Frequency (RF) signal couplers and one or more power couplers can be disposed in the receiver of the cradle. The one or more RF signal couplers can be configured to couple one or more RF communication signals to the wireless user device, while the one or more power couplers can be configured to couple power to the wireless user device. The coupling of power to the user device can be reduced or minimized when a downlink signal is received by a user device in the user device cradle, or when the user device cradle is in a weak signal area.

Abstract: A method and a device for image processing are disclosed. The method includes: receiving a first image, wherein the first image includes a face; detecting the face and a background region in the first image, establishing a three-dimensional model of the face according to the first image; rotating the three-dimensional model of the face by a first angle; and projecting the three-dimensional model of the face rotated by the first angle to an image coordinate system of the first image and fusing a face region with a processed background region to obtain a second image.