Abstract: Systems and methods for delivering cross-site auto-play media are described herein. The server can receive, from a client device, a request for media content at a first domain embedded in a content element from a second domain. The request can include an identifier indicating that the client device is configured to restrict cross-domain redirection. The server can determine, responsive to identifying the identifier indicating that the client device is configured to restrict cross-domain redirection, that the request comprises a content type header having a first predetermined value. The server can generate, responsive to the determination, a response comprising the media content element in a body of the response. The server can transmit the response to the client device. Receipt of the response can cause the client device to extract the media content element from the body of the response and to render the media content element.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a policy neural network used to select actions to be performed by a reinforcement learning agent interacting with an environment. In one aspect, a method includes obtaining path data defining a path through the environment traversed by the agent. A consistency error is determined for the path from a combined reward, first and last soft-max state values, and a path likelihood. A value update for the current values of the policy neural network parameters is determined from at least the consistency error. The value update is used to adjust the current values of the policy neural network parameters.

Abstract: In some implementations, a substrate for coupling to an integrated circuit includes multiple layers. Each of the multiple layers has, in a particular region of the substrate, a repeating pattern of regions corresponding to power and ground. The multiple layers include (i) a top layer having, in the particular region, power contacts and ground contacts for coupling to an integrated circuit and (ii) a bottom layer having, in the particular region, power contacts and ground contacts for coupling to another device. At least one layer of the multiple layers has a repeating pattern of signal traces that extend along and are located between the regions corresponding to ground in the at least one layer.

Abstract: Systems and methods are disclosed for incorporating user content within a communication session interface. In one implementation, a processing device receives, in relation to a communication session between a first user and a second user, a communication input associated with a first user. The processing device processes the communication input to identify one or more content items stored on a content sharing platform that are associated with the first user and that pertain to one or more aspects of the communication input. The processing device modifies one or more aspects of an interface of the communication session based on the one or more content items.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for distributing content items. In one aspect, a method includes identifying a content item that is eligible to be provided in response to a content item request and is associated with a bid. It is determined that the identified content item includes two or more different interactive elements that each cause different actions to be initiated in response to user interaction with the different interactive elements. An auction score is determined for the content item based on a function of the bid, a bid modifier for each different interaction, and a probability that each interaction will be invoked. The content item is selected to be provided based on the auction score. Data that cause presentation of the selected content item at a user device are output.

Abstract: The present disclosure provides an image capture, curation, and editing system that includes a resource-efficient mobile image capture device that continuously captures images. The mobile image capture device is operable to input an image into at least one neural network and to receive at least one descriptor of the desirability of a scene depicted by the image as an output of the at least one neural network. The mobile image capture device is operable to determine, based at least in part on the at least one descriptor of the desirability of the scene of the image, whether to store a second copy of such image and/or one or more contemporaneously captured images in a non-volatile memory of the mobile image capture device or to discard a first copy of such image from a temporary image buffer without storing the second copy of such image in the non-volatile memory.

Abstract: Methods for receiving a request to process, on a hardware circuit, a neural network comprising a first convolutional neural network layer having a stride greater than one, and in response, generating instructions that cause the hardware circuit to, during processing of an input tensor, generate a layer output tensor equivalent to an output of the first convolutional neural network layer by processing the input tensor using a second convolutional neural network layer having a stride equal to one but that is otherwise equivalent to the first convolutional neural network layer to generate a first tensor, zeroing out elements of the first tensor that would not have been generated if the second convolutional neural network layer had the stride of the first convolutional neural network layer to generate a second tensor, and performing max pooling on the second tensor to generate the layer output tensor.

Abstract: A device is provided that includes (a) an antenna that includes at least one conductor, (b) a housing that includes an inner-upper surface and an inner-lower surface separated by a first distance, (c) a battery disposed within the housing, where a base surface of the battery is proximate to the inner-lower surface of the housing, where a first portion of the battery has a height, which is substantially equal to the first distance, and where a second portion of the battery is of lesser height than the first portion of the battery such that space exists between the second portion of the battery and the inner-upper surface of the housing, and (d) where the one conductor is arranged over the second portion of the battery in the space, such that the one conductor and the battery do not contact one another, and where, as arranged, the antenna is capable of a far-field communication.

Abstract: Systems and methods for selecting content based on an event associated with a device identifier are provided. One or more processors can receive a request to serve content. The processors can identify a device identifier associated with the request. The processors can determine, from the device identifier, an event for which to serve content. The processors can determine, from the request, a length of time between a time the request to serve content is received and a time at which the event is scheduled to occur. The processors can select, based on the determined length of time and event parameters associated with the event, content for display and provide the selected content for display at a computing device associated with the device identifier.

Abstract: Systems and methods herein allow for automatic monitoring of conversions associated with uniform resource identifiers (URIs) accessed via client applications but actuated on web browsers. Upon actuation of a URI on a browser of a client device, a data processing system can assign a click ID to the actuation event and send it to the client device. The client device can open the URI in a client application and send the click ID and a client ID associated with the client device to the client device. The data processing system can map the click ID to the client ID. Upon occurrence of a URI related conversion, on the client application, the data processing system can receive a conversion message indicative of the conversion and including the client ID, and link the URI related conversion to the click ID based on the stored mapping.

Abstract: A system including a digital component request server that receives digital component requests for a supplemental digital component that will be presented in an additional multicast stream that is different to a multicast stream that is delivering content; a digital component distribution server that i) selects supplemental digital components to distribute over the additional multicast streams, the supplemental digital components being selected for one of the additional multicast streams being different to the supplemental digital component selected for each of the other additional multicast streams, and that ii) assigns one group of a plurality of groups of the multiple client devices to a respective one of the additional multicast streams; and an instruction generation server that generates instructions for the multiple client devices to tune to the respective additional multicast streams assigned to the group of multiple client devices that includes the different client device that submitted the request.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for acoustic modeling of audio data. One method includes receiving audio data representing a portion of an utterance, providing the audio data to a trained recurrent neural network that has been trained to indicate the occurrence of a phone at any of multiple time frames within a maximum delay of receiving audio data corresponding to the phone, receiving, within the predetermined maximum delay of providing the audio data to the trained recurrent neural network, output of the trained neural network indicating a phone corresponding to the provided audio data using output of the trained neural network to determine a transcription for the utterance, and providing the transcription for the utterance.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a model using parameter server shards. One of the methods includes receiving, at a parameter server shard configured to maintain values of a disjoint partition of the parameters of the model, a succession of respective requests for parameter values from each of a plurality of replicas of the model; in response to each request, downloading a current value of each requested parameter to the replica from which the request was received; receiving a succession of uploads, each upload including respective delta values for each of the parameters in the partition maintained by the shard; and updating values of the parameters in the partition maintained by the parameter server shard repeatedly based on the uploads of delta values to generate current parameter values.

Abstract: A method includes receiving data corresponding one of streaming data or batch data and a content of the received data for computation. The method also includes determining an event time of the data for slicing the data, determining a processing time to output results of the received data, and emitting at least a portion of the results of the received data based on the processing time and the event time.

Abstract: The technology employs a micro-scheduler arrangement to detect and correct issues relating to software code testing. The micro-scheduler uses test results from unsubmitted branches to predict test breakages, test fixes, and to identify which changes have broken certain tests. This arrangement is high beneficial in large-scale software systems that have millions or billions of lines of code, and can be used to supplement an existing test automation platform. In response to receipt of a new presubmit or postsubmit test result, the system evaluates the recent test history in relation to milestone tests in order to determine whether or not to schedule a particular test for one or more change list identifiers. Different test conditions, such as currently failing and currently broken, in conjunction with stored presubmit information, help the system determine the range of changes that could be the culprit for a test breakage.

Abstract: This disclosure generally relates to systems and methods that facilitate employing a server based content recording component for recording content remotely for a client content video recording device that has limited storage and limited number of content streams to which it can access concurrently, while satisfying content rules for broadcast and re-transmission.

Abstract: An image processor is described. The image processor includes an I/O unit to read input image data from external memory for processing by the image processor and to write output image data from the image processor into the external memory. The I/O unit includes multiple logical channel units. Each logical channel unit is to form a logical channel between the external memory and a respective producing or consuming component within the image processor. Each logical channel unit is designed to utilize reformatting circuitry and addressing circuitry. The addressing circuitry is to control addressing schemes applied to the external memory and reformatting of image data between external memory and the respective producing or consuming component. The reformatting circuitry is to perform the reformatting.