Abstract: Techniques and apparatuses are described for video frame codec architectures. A frame decompressor decompresses compressed frames to produce decompressed frames. A frame decompressor controller arbitrates shared access to the frame decompressor. Multiple cores of an SoC request to receive a decompressed frame from the frame decompressor via the frame decompressor controller. The frame decompressor controller can implement a request queue and can order the servicing of requests based on priority of the requests or requesting cores. The frame decompressor controller can also establish a time-sharing protocol for access by the multiple cores. In some implementations, a video decoder is logically integrated with the frame decompressor and stores portions of a decompressed frame in a video buffer, and a display controller retrieves the portions for display using a synchronization mechanism. In analogous manners, a frame compressor controller can arbitrate shared access to a frame compressor for the multiple cores.

Abstract: A method including at each of a number of client devices receiving a data item, receiving a public key from a second computing system, encrypting the data item using the public key to produce a singly encrypted data item, engaging in an oblivious pseudorandom function protocol with a first computing system using the singly encrypted data item to produce a seed, generating an encrypted secret share using a threshold secret sharing function under which the encrypted secret share cannot be decrypted until a threshold number of encrypted secret shares associated with the same singly encrypted data item are received, and transmitting the encrypted secret share to the first computing system and at the first computing system receiving a number of encrypted secret shares from the number of client devices, processing the number of encrypted secret shares to produce processed data, and transmitting the processed data to a second computing system.

Abstract: This document describes systems and techniques directed at mitigating display diffraction flares for under-display sensing. In aspects, an equation may be derived that models the effects of a display in producing a diffraction phenomenon at an image plane of a sensing region for an under-display light-sensing device. The equation may be used to determine an arrangement (e.g., an optimized arrangement) of components (e.g., sub-pixels) within the display that minimizes a diffraction efficiency for at least one diffraction order and, thereby, mitigates an intensity and/or a prevalence of optical artifacts in light-sensing data. In implementations, an image intensity point-spread-function is utilized to calculate diffraction efficiencies for respective diffraction orders (e.g., the lowest diffraction orders, the diffraction orders with the greatest brightness).

Abstract: Various arrangements for short-range wireless communication between audio output devices, such as true wireless earbuds, are presented herein. A first earbud of a pair of earbuds may determine that a first audio packet addressed to the first earbud from an audio source was not properly received. However, a second earbud of the pair of earbuds may properly receive the first audio packet addressed to the first earbud. The second earbud can then, directly to the first earbud, transmit a cross acknowledgement indicating that the second earbud properly received the audio packet.

Abstract: Various arrangements for short-range wireless communication are presented herein. An earbud of a pair of true wireless earbuds can receive an audio packet addressed to the other earbud of the pair. A single connected isochronous stream (CIS) within a connected isochronous group (CIG) may be present between the pair of true wireless earbuds and an audio source which transmitted the audio packet. The earbud can transmit a cross-acknowledgement indicating receipt of the audio packet to the other earbud. The earbud can also transmit audio data from the audio packet to the other earbud after the cross acknowledgement.

Abstract: A method for receiving processed information at a remote device is described. The method includes transmitting from the remote device a verbal request to a first information provider and receiving a digital message from the first information provider in response to the transmitted verbal request. The digital message includes a symbolic representation indicator associated with a symbolic representation of the verbal request and data used to control an application. The method also includes transmitting, using the application, the symbolic representation indicator to a second information provider for generating results to be displayed on the remote device.

Abstract: Computer-implemented techniques can include obtaining, by a client computing device, a digital media item and a request for a processing task on the digital item and determining a set of operating parameters based on (i) available computing resources at the client computing device and (ii) a condition of a network. Based on the set of operating parameters, the client computing device or a server computing device can select one of a plurality of artificial neural networks (ANNs), each ANN defining which portions of the processing task are to be performed by the client and server computing devices. The client and server computing devices can coordinate processing of the processing task according to the selected ANN. The client computing device can also obtain final processing results corresponding to a final evaluation of the processing task and generate an output based on the final processing results.

Abstract: A method for using a user-fillable form in a host container includes receiving, at a host container, a user-fillable form bound to dynamic data from an underlying data source where the user-fillable form has a data structure generated by prepopulated coding. The method further includes translating the user-fillable form into a hostable format for the host container. The method also includes rendering, using the hostable format for the host container, the user-fillable form in a user interface. The method further includes receiving, at the user interface of the host container, from a user of the host container, a data entry for input to the user-fillable form and updating, by the host container, the dynamic data from the underlying data source by persisting data from the data entry in a data store associated with the underlying data source.

Abstract: A method includes detecting multiple users, receiving a first query issued by a first user, the first query including a command for a digital assistant to perform a first action, and enabling a round robin mode to control performance of actions commanded by queries. The method also includes, while performing the first action, receiving audio data corresponding to a second query including a command to perform a second action, performing speaker identification on the audio data, determining that the second query was spoken by the first user, preventing performing the second action, and prompting at least another user to issue a query. The method further includes receiving a third query issued by a second user, the third query including a command for the digital assistant to perform a third action, and when the digital assistant completes performing the first action, executing performance of the third action.

Abstract: Various arrangements of wireless earbuds are presented. A first earbud, can include a first speaker, a first processing system, and a first wireless communication interface, that communicates with an audio source device using Bluetooth communications. A second earbud can include a second speaker, a second processing system, and a second wireless communication interface, that communicates with the audio source device and the first earbud using Bluetooth communications. The first earbud and the second earbud may be configured to wirelessly communicate with each other following completion of a first connected isochronous stream (CIS) event for the first earbud and second CIS event for the second earbud within a connected isochronous group (CIG) event.

Abstract: A method for handling contradictory queries on a shared device includes receiving a first query issued by a first user, the first query specifying a first long-standing operation for a digital assistant to perform, and while the digital assistant is performing the first long-standing operation, receiving a second query, the second query specifying a second long-standing operation for the digital assistant to perform. The method also includes determining that the second query was issued by another user different than the first user and determining, using a query resolver, that performing the second long-standing operation would conflict with the first long-standing operation. The method further includes identifying one or more compromise operations for the digital assistant to perform, and instructing the digital assistant to perform a selected compromise operation among the identified one or more compromise operations.

Abstract: A method for an aggregatable application programming interface (API) includes receiving, from a third party service, an aggregation request requesting aggregation of client data from a client of the third party service. The method also includes receiving, from an API executed by a client device of the client, a first portion of the client data. The method includes storing the first portion of the client data and receiving, from the API, a second portion of the client data. The method includes determining that the second portion of the client data is a final portion of the client data. In response, the method includes aggregating the first portion of the client data with the second portion of the client data. The method also includes transmitting the aggregated client data to the third party service.

Abstract: Systems and methods for video conference content auto-retrieval and focus based on learned relevance is provided. In accordance with the systems and methods, audio streams and video streams from client devices participating in a video conference are received. Based on the audio streams, a subject being discussed during the video conference at a point in time is determined. A video stream that is most relevant to the subject being discussed during the video conference at the point in time is determined from the video streams. The determined video stream is provided to the client devices for presentation on the client devices while the subject is being discussed during the video conference.

Abstract: An apparatus for encoding an image block includes a processor that presents, to a machine-learning model, the image block, obtains the partition decision for encoding the image block from the model, and encodes the image block using the partition decision. The model is trained to output a partition decision for encoding the image block by using training data for a plurality of training blocks as input, the training data including for a training block, partition decisions for encoding the training block, and, for each partition decision, a rate-distortion value resulting from encoding the training block using the partition decision. The model is trained using a loss function combining a partition loss function based upon a relationship between the partition decisions and respective predicted partitions, and a rate-distortion cost loss function based upon a relationship between the rate-distortion values and respective predicted rate-distortion values.

Abstract: A User Equipment 210 (UE) determines 215 a maximum UL/DL duty cycle to utilize for a connection with a base station 202 based on a content of a UE capabilities request 212 sent by the base station, where the maximum UL/DL duty cycle is typically is less than a most restrictive, possible UL/DL duty cycle of the UE, yet allows the UE to remain SAR compliant. The UE may determine 215 the maximum UL/DL duty cycle further based on, e.g., requested frequency bands, operational UE transceivers, carrier components, UE power class, stored power data, etc. The UE communicates 218 the maximum UL/DL duty cycle for the connection to the base station, thereby resulting in more efficient downlink data delivery during the connection as well as increasing cell site coverage and overall system efficiency. The UE may store a global parameter whose value is indicative of the maximum UL/DL duty cycle.

Abstract: Methods, systems, and media for identifying videos containing objectionable content are provided.

Abstract: A device includes: a first chip including a qubit; and a second chip bonded to the first chip, the second chip including a substrate including first and second opposing surfaces, the first surface facing the first chip, wherein the second chip includes a single layer of superconductor material on the first surface of the substrate, the single layer of superconductor material including a first circuit element. The second chip further includes a second layer on the second surface of the substrate, the second layer including a second circuit element. The second chip further includes a through connector that extends from the first surface of the substrate to the second surface of the substrate and electrically connects a portion of the single layer of superconducting material to the second circuit element.

Abstract: Embodiments are provided for communicating notifications and other textual data associated with applications installed on an electronic device. According to certain aspects, a user can interface with an input device to send (218) a wake up trigger to the electronic device. The electronic device retrieves (222) application notifications and converts (288) the application notifications to audio data. The electronic device also sends (230) the audio data to an audio output device for annunciation (232). The user may also use the input device to send (242) a request to the electronic device to activate the display screen. The electronic device identifies (248) an application corresponding to an annunciated notification, and activates (254) the display screen and initiates the application.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for hotword detection on multiple devices are disclosed. In one aspect, a method includes the actions of receiving, by a computing device, audio data that corresponds to an utterance. The actions further include determining a likelihood that the utterance includes a hotword. The actions further include determining a loudness score for the audio data. The actions further include based on the loudness score, determining an amount of delay time. The actions further include, after the amount of delay time has elapsed, transmitting a signal that indicates that the computing device will initiate speech recognition processing on the audio data.

Abstract: Aspects of the disclosure are directed to monitoring, alerting, and/or root causing network problems based on current network behavior and network events at any instant in time using a network behavior model. The network behavior model can learn and be updated with network states and events to correlate network problems and determine root causes of the network problems for alerting and/or automatic correction.

Abstract: A method of phrase extraction for ASR models includes obtaining audio data characterizing an utterance and a corresponding ground-truth transcription of the utterance and modifying the audio data to obfuscate a particular phrase recited in the utterance. The method also includes processing, using a trained ASR model, the modified audio data to generate a predicted transcription of the utterance, and determining whether the predicted transcription includes the particular phrase by comparing the predicted transcription of the utterance to the ground-truth transcription of the utterance. When the predicted transcription includes the particular phrase, the method includes generating an output indicating that the trained ASR model leaked the particular phrase from a training data set used to train the ASR model.

Abstract: This document generally describes a neural network training system, including one or more computers, that trains a recurrent neural network (RNN) to receive an input, e.g., an input sequence, and to generate a sequence of outputs from the input sequence. In some implementations, training can include, for each position after an initial position in a training target sequence, selecting a preceding output of the RNN to provide as input to the RNN at the position, including determining whether to select as the preceding output (i) a true output in a preceding position in the output order or (ii) a value derived from an output of the RNN for the preceding position in an output order generated in accordance with current values of the parameters of the recurrent neural network.

Abstract: The present disclosure provides systems and methods for content quasi-personalization or anonymized content retrieval via aggregated browsing history of a large plurality of devices, such as millions or billions of devices. A sparse matrix may be constructed from the aggregated browsing history, and dimensionally reduced, reducing entropy and providing anonymity for individual devices. Relevant content may be selected via quasi-personalized clusters representing similar browsing histories, without exposing individual device details to content providers.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: A method for pervasive resource identification includes receiving an authentication request from a first application service. The authentication request requests authentication of a user of a user device. The method includes obtaining device information associated with the user device of the user and generating a unique opaque identifier for the user device based on the device information. The method includes obtaining authentication credentials from the user device. The authentication credentials verify an identity of the user. In response to receiving the authentication credentials from the user device, the method includes generating an authentication token and encoding the unique opaque identifier into the authentication token. The method also includes transmitting the authentication token to the first application service.

Abstract: The technology generally relates to determining a status of an optical channel between two components. For example, the components may be connected via an optical link including a plurality of optical channels. A first portion of the optical channels may be in use such that a second portion of the optical channels may be redundant channels. The component may include a test generator that transmits and receives a data pattern over each channel. The test generator may determine, based on the received data pattern, a status of each of the channels. If the status of a given channel is a failure status, the component may divert data for the given channel to a redundant channel.

Abstract: This document describes techniques and apparatuses for a user-equipment-coordination set for disengaged mode. In aspects, a base station forms a disengaged-mode-user-equipment coordination set including multiple user equipment operating in a disengaged mode. The disengaged-mode-user-equipment coordination set uses joint transmission and reception to communicate with the base station. The base station communicates control-plane information to an individual user equipment or multiple user equipment in the disengaged-mode-user-equipment coordination set.

Abstract: The present disclosure is directed to systems and methods for performing reading comprehension with machine learning. More specifically, the present disclosure is directed to a Neural Symbolic Reader (example implementations of which may be referred to as NeRd), which includes a reader to encode the passage and question, and a programmer to generate a program for multi-step reasoning. By using operators like span selection, the program can be executed over a natural language text passage to generate an answer to a natural language text question. NeRd is domain-agnostic such that the same neural architecture works for different domains. Further, NeRd is compositional such that complex programs can be generated by compositionally applying the symbolic operators.

Abstract: The present technology proposes techniques for generating globally coherent timestamps. This technology may allow distributed systems to causally order transactions without incurring various types of communication delays inherent in explicit synchronization. By globally deploying a number of time masters that are based on various types of time references, the time masters may serve as primary time references. Through an interactive interface, the techniques may track, calculate and record data relative to each time master thus providing the distributed systems with causal timestamps.

Abstract: A heat distribution device comprising a main body, a recessed cavity positioned within the main body, the recessed cavity having an interior surface, a peripheral wall extending around and defining the interior surface, and a central point within the recessed cavity. A plurality of ribs may extend away from the interior surface of the recessed cavity. The plurality of ribs may be concentrically arranged around the central point and define a plurality of channels therebetween. Each of the plurality of ribs may have a top surface that slopes toward or away from the central point. The plurality of ribs may be arranged so that the top surfaces of the plurality of ribs collectively form a collective sloped surface within the heat distribution device.

Abstract: A system enables universal remote media control across multiple platforms, devices, and users. A protocol allows a cast controller to have access to media operations and a state(s) associated with media content. The system receives commands from a cast controller, provides the commands to a media player, loads new media content into the media player, based on the commands, and receives state notifications from the media player. Another cast controller can receive the state notifications from the media player and control the media player based on the media operations and state(s) associated with media content.

Abstract: System and methods for displaying one or more assets on a client device based on device characteristics are provided. Code is transmitted to a client device. The code, when executed by the client device, causes a processor of the client device to determine a first device characteristic of the client device. A first layout may be selected based on the first device characteristic. The layout may include one or more cards. Each card may correspond to one or more assets. Each card may be modified based on a corresponding card characteristic. One or more assets may be requested. The assets may be displayed on the first client device. The code may be transmitted to another client device, which may select a different layout based on a different device characteristic.

Abstract: Techniques for presenting a search result with an improved user interface. A computer system can receive, from a user device, a request for a content item. Additionally, the system can select, based on the request, a first content item from a plurality of content items. The first content item can be associated with an organization image and an organization name of an organization. Moreover, the system can process, using one or more machine-learned model, the organization image to determine whether the organization image is acceptable to be presented in the search result. Subsequently, the system can transmit, to the user device, the first content item and the organization image to be presented in the search result.

Abstract: A computer-implemented method, computer program product, and computing system for rendering an annotatable image within an image viewer. An indication of an intent to annotate the annotatable image is received from a user. A meme generation interface is rendered with respect to the annotatable image. Meme annotation criteria is received from the user via the meme generation interface. The meme annotation criteria includes one or more of: a meme message, a meme position indicator, and a font type identifier. The annotatable image is modified based, at least in part, upon the meme annotation criteria, thus generating an annotated image. The annotated image is published to a meme publication website.

Abstract: A system for automated guided vehicle safety may include an automated guided vehicle (AGV) having a propulsion system configured to move the AGV, and a processor configured to control the propulsion system, and a laser imaging system configured to deploy a virtual safety fence at least partially surrounding the AGV. The laser imaging system may include a plurality of laser imaging sensors including a front sensor and a rear sensor, and a movable boom, the front sensor being mounted to the movable boom and configured to extend in front of the housing of the AGV.

Abstract: Systems and methods for creating a live copy of a data object from a production system for use by third party applications. The systems and methods include receiving at least one request for a copy of production data from an application; creating a live backup copy; creating a flash copy of the live backup copy, and a flash copy bitmap; creating a modified version of the live backup copy by changing a subset of data in the live backup copy; recording the changed subset of data using the flash copy bitmap; mounting, the modified version of the live backup copy to the application; and transforming the modified version of the live backup copy back to the live backup copy when unmounting the modified version of the live backup copy of the production data from the application by applying changes associated with the flash copy bitmap to the live backup copy.

Abstract: Methods, systems, and media for generating and evaluating street grids comprising: receiving street grid information corresponding to a plurality of locations, wherein the street grid information corresponding to a location is associated with vehicular traffic information; training a pedestrian comfort model using the street grid information and the vehicular traffic information, wherein an output of the pedestrian comfort model is a predicted pedestrian comfort score that is based on traffic congestion from the vehicular traffic information; receiving a plurality of potential street grids; evaluating each potential street grid in the plurality of potential street grids using the trained pedestrian comfort model, wherein the trained pedestrian comfort model generates predicted pedestrian comfort scores for portions of each potential street grid; and generating an augmented map of each potential street grid that presents the predicted pedestrian comfort scores for each portion of each potential street grid.

Abstract: The technology involves providing haptic feedback to a user of a computing device. For instance, user input associated with a program of the computing device is received by a user interface module of the computing device. One or more processors of the computing device determine a kind of interaction corresponding to the user input. The one or more processors identify whether the kind of interaction is associated with one or more haptic feedback effects of a curated suite of haptic effects. Upon identifying that the kind of interaction is associated with one or more haptic feedback effect, the one or more processors select a particular haptic feedback effect from the curated suite of haptic effects. And the one or more processors are able to cause a haptic feedback module of the computing device to provide the particular haptic feedback effect for sensation by the user.

Abstract: An IC die includes a temperature control element suitable for three-dimensional IC package with enhanced thermal control and management. The temperature control element may assist temperature control of the IC die when in operation. In one example, the temperature control element may have a plurality of thermal dissipating features disposed on a first surface of the IC die to efficiently control and dissipate the thermal energy from the IC die when in operation. A second surface opposite to the first surface of the IC die may include a plurality of devices, such as semiconductors transistors, devices, electrical components, circuits, or the like, that may generate thermal energy when in operation. The temperature control element may provide an IC die with high efficiency of heat dissipation that is suitable for 3D IC package structures and requirements.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a generator neural network to adapt input images.

Abstract: In one exemplary aspect, the subject matter described in this specification can be embodied in an energy extraction system that includes a decelerator cavity coupled to a transport line for a charged particle beam and an energy conversion device coupled to the decelerator cavity. The decelerator cavity is configured to extract energy from the charged particle beam traveling through the decelerator cavity as RF energy. The energy conversion is configured to convert the RF energy into electrical current and supply the electrical current to an electric power grid. The charged particle beam includes charged particles with individual rest masses greater than the rest mass of an electron.

Abstract: Churn-aware training of a classifier which reduces the difference between predictions of two different models, such as a prior generation of a classification model and a subsequent generation. A second dataset of labelled data is scored on a prior generation of a classification model, wherein the prior generation was trained on a first dataset of labelled data. A subsequent generation of a classification model is trained with the second dataset of labelled data, wherein in training of the subsequent generation, weighting of at least some of the labelled data in the second dataset, such as labelled data threat yielded an incorrect classification, is adjusted based on the score of such labelled data in the prior generation.

Abstract: A method includes receiving data objects, determining a predicted lifespan of each data object, and instantiating multiple shard files. Each shard file has an associated predicted lifespan range. The method also includes writing each data object into a corresponding shard file having the associated predicted lifespan range that includes the predicted lifespan of the respective data object and storing the shard files in a distributed system. The method also includes determining whether any stored shard files satisfy a compaction criteria based on a number of deleted data objects in each corresponding stored shard file. For each stored shard file satisfying the compaction criteria, the method also includes compacting the stored shard file by rewriting the remaining data objects of the stored shard file into a new shard file.

Abstract: Systems and methods for training models to predict dense correspondences across images such as human images. A model may be trained using synthetic training data created from one or more 3D computer models of a subject. In addition, one or more geodesic distances derived from the surfaces of one or more of the 3D models may be used to generate one or more loss values, which may in turn be used in modifying the model's parameters during training.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a final classification output for an image of eye tissue. The image is provided as input to each of one or more segmentation neural networks to obtain one or more segmentation maps of the eye tissue in the image. A respective classification input is generated from each of the segmentation maps. For each of the segmentation maps, the classification input for the segmentation map is provided as input to each of one or more classification neural networks to obtain, for each segmentation map, a respective classification output from each classification neural network. A final classification output for the image is generated from the respective classification outputs for each of the segmentation maps.

Abstract: An example method includes estimating, based on content to be displayed at a display of a mobile computing device at a future time, an amount of power to be used by the display at the future time; selecting, based on the estimated power level, a power converter of a plurality of power converters of the mobile computing device, each of the plurality of power converters optimized for a different output power range; and causing electrical power from the selected power converter to be supplied to the display at the future time.

Abstract: A system according to an embodiment of the present invention includes at least two virtual machines running on a hardware platform using either a hosted or a bare metal hypervisor. The virtual machines may communicate with an agent-server resident in the host operating system or in one of the virtual machines to switch control of the hardware component, such as graphics hardware, from one virtual machine to another.

Abstract: A method for a crawl algorithm includes obtaining a plurality of web pages for a web crawler to crawl. The method also includes determining an available bandwidth for the web crawler. The method includes, for each respective web page of the plurality of web pages, determining a respective crawl value for the respective web page based on the available bandwidth and determining that the respective crawl value of the respective web page satisfies a threshold value. The method includes, in response to determining that the respective crawl value of the respective web page satisfies the threshold value, updating the respective web page in a cache memory.

Abstract: This document describes techniques and devices for motion-induced blurring to reduce an appearance of a boundary separating regions of a display. Sensors (e.g., cameras, microphones, biometric sensors, ambient light sensors, radar sensors, and so forth) may be placed at least partially underneath regions of a display. Placing a sensor underneath a region of a display, however, often requires reduced pixel-density (e.g., relatively low resolution), which can cause scintillations of the content as it moves on the display. The techniques described herein address some undesirable effects of this reduced pixel-density by blurring content moving within regions of low resolution. Furthermore, the techniques can adjust an amount of blurring based on a speed of the moving content. Thus, when the display includes regions of differing resolutions, the techniques described herein can blur the moving content to reduce the appearance of a boundary separating these regions of differing resolutions.