Abstract: The present disclosure relates to systems and methods for operating transceiver circuitry to transmit or receive signals on various frequency ranges. To do so, an electronic device may determine a receive delay between one or more messages received on different component carriers and may transmit the receive delay to a base station to update how communications are transmitted on one of the component carriers. The update made to at least one of the component carriers may compensate for the receive delay between the different component carriers. Compensating for the receive delay may improve operations that delay downlink communications to reduce a likelihood or stop simultaneous downlink and uplink communications by further adjusting for delays seen at an electronic device when communicating with base stations disposed at a different distances from the electronic device.

Abstract: The present disclosure relates to systems and methods for operating transceiver circuitry to communicate signals on various frequency ranges based on primary cell and/or secondary cell assignments. A controller associated with the transceiver circuitry may operate a receiver to receive a handover initiation packet using a first frequency range as a primary cell from a base station. The controller may adjust the receiver and a transmitter to use a second frequency range as the primary cell, to adjust the receiver and the transmitter to detach from the first frequency range as the primary cell, and to reset information corresponding to the first frequency range to enable attachment of the first frequency range to a secondary cell.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes a batch normalization layer. One of the methods includes receiving a respective first layer output for each training example in the batch; computing a plurality of normalization statistics for the batch from the first layer outputs; normalizing each component of each first layer output using the normalization statistics to generate a respective normalized layer output for each training example in the batch; generating a respective batch normalization layer output for each of the training examples from the normalized layer outputs; and providing the batch normalization layer output as an input to the second neural network layer.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes a batch normalization layer. One of the methods includes receiving a respective first layer output for each training example in the batch; computing a plurality of normalization statistics for the batch from the first layer outputs; normalizing each component of each first layer output using the normalization statistics to generate a respective normalized layer output for each training example in the batch; generating a respective batch normalization layer output for each of the training examples from the normalized layer outputs; and providing the batch normalization layer output as an input to the second neural network layer.

Abstract: A neural network system that includes: multiple subnetworks that includes: a first subnetwork including multiple first modules, each first module including: a pass-through convolutional layer configured to process the subnetwork input for the first subnetwork to generate a pass-through output; an average pooling stack of neural network layers that collectively processes the subnetwork input for the first subnetwork to generate an average pooling output; a first stack of convolutional neural network layers configured to collectively process the subnetwork input for the first subnetwork to generate a first stack output; a second stack of convolutional neural network layers that are configured to collectively process the subnetwork input for the first subnetwork to generate a second stack output; and a concatenation layer configured to concatenate the pass-through output, the average pooling output, the first stack output, and the second stack output to generate a first module output for the first module.

Abstract: Systems and methods are disclosed for coordinating transmission and reception of data according to multiple communication standards over a frequency band. In particular, one or more base stations/mobile electronic devices may determine a first data size of first data to be sent conforming to a first communication standard and a second data size of second data conforming to a second communication standard. A first time period may then be determined for which to send the first data based on the first data size, and a second time period may be determined for which to send the second data based on the second data size. In response to determining that the frequency channel is clear of other transmissions, the first data may be sent according to the first standard in the first time period, and the second data may be sent according to the second standard in the second time period.

Abstract: An electronic device discussed herein may include radio frequency communication circuitry for communication on a radio frequency network according to a communication configuration, a processor, and memory. The memory may store instructions that, when executed by the processor, cause the electronic device to perform operations including receiving, a first muting configuration indicating when the radio frequency communication circuitry is to communicate using a first type of communication on a first frequency band and when the radio frequency communication circuitry is to communicate using a second type of communication on a second frequency band, where the first frequency band may overlap with the second frequency band. The memory may store instructions that, when executed by the processor, cause the electronic device to perform operations including transmitting or receiving a data packet using the radio frequency communication circuitry according to the communication configuration.

Abstract: Aspects of the present disclosure are directed to novel activation functions which enable improved reproducibility and accuracy tradeoffs in neural networks. In particular, the present disclosure provides a family of activation functions that, on one hand, are smooth with continuous gradient and optionally monotonic but, on the other hand, also mimic the mathematical behavior of a Rectified Linear Unit (ReLU). As examples, the activation functions described herein include a smooth rectified linear unit function and also a leaky version of such function. In various implementations, the proposed functions can provide both a complete stop region and a constant positive gradient (e.g., that can be 1) pass region like a ReLU, thereby matching accuracy performance of a ReLU. Additional implementations include a leaky version and/or functions that feature different constant gradients in the pass region.

Abstract: An electronic device discussed herein may communicatively couple to a base station. The base station may receive a first paging cycle assignment corresponding to a first subscriber identification module (SIM) card and determine a second paging cycle assignment for use with a second SIM card. The second paging cycle assignment may be generated based on the first paging cycle assignment. The base station may communicate with the electronic device using the second paging cycle assignment. The second paging cycle assignment may guide the base station to transmit data to the electronic device without interrupting a transmission made according to the first paging cycle assignment.

Abstract: Methods, systems, and media for summarizing a video with video thumbnails are provided.

Abstract: Systems and methods for using ranging signals with cellular devices. The ranging signals may utilize ranging slots and resources at least partially allocated by a cellular network. The resources may include frequencies used for uplink or downlink communications between the cellular network and the cellular devices. Alternatively, the resources may include frequencies outside of a spectrum used for communication between the cellular network and the cellular devices.

Abstract: When user equipment (UE) is to be handed over, the network and/or the UE determines a best beam for the UE's interactions with the target cell before the handover is completed. One or more additional next best beams may also be determined. The network (e.g., the target cell) allocates one or more uplink (UL) grants that corresponds to the best beam. Via a current cell, the UE receives the one or more UL grants from the network pertaining to communications between the UE and the target cell. The UE determines whether any beams of the one or more UL grants satisfy beam criteria. The beam criteria may include 1) an allocated beam being the current best beam or 2) an allocated beam being within a strength threshold of the current best beam. If the criteria is not satisfied, the UE initiates another handover type (e.g., a RACH-based handover).

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes a batch normalization layer. One of the methods includes receiving a respective first layer output for each training example in the batch; computing a plurality of normalization statistics for the batch from the first layer outputs; normalizing each component of each first layer output using the normalization statistics to generate a respective normalized layer output for each training example in the batch; generating a respective batch normalization layer output for each of the training examples from the normalized layer outputs; and providing the batch normalization layer output as an input to the second neural network layer.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing a neural network. In one aspect, the neural network includes a batch renormalization layer between a first neural network layer and a second neural network layer. The first neural network layer generates first layer outputs having multiple components. The batch renormalization layer is configured to, during training of the neural network on a current batch of training examples, obtain respective current moving normalization statistics for each of the multiple components and determine respective affine transform parameters for each of the multiple components from the current moving normalization statistics. The batch renormalization layer receives a respective first layer output for each training example in the current batch and applies the affine transform to each component of a normalized layer output to generate a renormalized layer output for the training example.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images or features of images using an image classification system that includes a batch normalization layer. One of the systems includes a convolutional neural network configured to receive an input comprising an image or image features of the image and to generate a network output that includes respective scores for each object category in a set of object categories, the score for each object category representing a likelihood that that the image contains an image of an object belonging to the category, and the convolutional neural network comprising: a plurality of neural network layers, the plurality of neural network layers comprising a first convolutional neural network layer and a second neural network layer; and a batch normalization layer between the first convolutional neural network layer and the second neural network layer.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing long-short term memory cells with saturating gating functions. One of the systems includes a first Long Short-Term Memory (LSTM) cell, wherein the first LSTM cell is configured to, for each of the plurality of time steps, generate a new cell state and a new cell output by applying a plurality of gates to a current cell input, a current cell state, and a current cell output, each of the plurality of gates being configured to, for each of the plurality of time steps: receive a gate input vector, generate a respective intermediate gate output vector from the gate input, and apply a respective gating function to each component of the respective intermediate gate output vector, wherein the respective gating function for at least one of the plurality of gates is a saturating gating function.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing a neural network. In one aspect, the neural network includes a batch renormalization layer between a first neural network layer and a second neural network layer. The first neural network layer generates first layer outputs having multiple components. The batch renormalization layer is configured to, during training of the neural network on a current batch of training examples, obtain respective current moving normalization statistics for each of the multiple components and determine respective affine transform parameters for each of the multiple components from the current moving normalization statistics. The batch renormalization layer receives a respective first layer output for each training example in the current batch and applies the affine transform to each component of a normalized layer output to generate a renormalized layer output for the training example.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images or features of images using an image classification system that includes a batch normalization layer. One of the systems includes a convolutional neural network configured to receive an input comprising an image or image features of the image and to generate a network output that includes respective scores for each object category in a set of object categories, the score for each object category representing a likelihood that that the image contains an image of an object belonging to the category, and the convolutional neural network comprising: a plurality of neural network layers, the plurality of neural network layers comprising a first convolutional neural network layer and a second neural network layer; and a batch normalization layer between the first convolutional neural network layer and the second neural network layer.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing images or features of images using an image classification system that includes a batch normalization layer. One of the systems includes a convolutional neural network configured to receive an input comprising an image or image features of the image and to generate a network output that includes respective scores for each object category in a set of object categories, the score for each object category representing a likelihood that that the image contains an image of an object belonging to the category, and the convolutional neural network comprising: a plurality of neural network layers, the plurality of neural network layers comprising a first convolutional neural network layer and a second neural network layer; and a batch normalization layer between the first convolutional neural network layer and the second neural network layer.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes a batch normalization layer. One of the methods includes receiving a respective first layer output for each training example in the batch; computing a plurality of normalization statistics for the batch from the first layer outputs; normalizing each component of each first layer output using the normalization statistics to generate a respective normalized layer output for each training example in the batch; generating a respective batch normalization layer output for each of the training examples from the normalized layer outputs; and providing the batch normalization layer output as an input to the second neural network layer.