Abstract: A system includes a neural network that includes a Mixture of Experts (MoE) subnetwork between a first neural network layer and a second neural network layer. The MoE subnetwork includes multiple expert neural networks. Each expert neural network is configured to process a first layer output generated by the first neural network layer to generate a respective expert output. The MoE subnetwork further includes a gating subsystem that selects, based on the first layer output, one or more of the expert neural networks and determine a respective weight for each selected expert neural network, provides the first layer output as input to each of the selected expert neural networks, combines the expert outputs generated by the selected expert neural networks in accordance with the weights for the selected expert neural networks to generate an MoE output, and provides the MoE output as input to the second neural network layer.

Abstract: Methods, systems and apparatus for implementing iSWAP quantum logic gates between a first qubit and a second qubit. In one aspect, a method includes implementing a cascade schedule that defines a trajectory of a detuning between a frequency of the first qubit and a frequency of the second qubit. Implementing the cascade schedule includes: during a first stage, adiabatically driving detuning between the frequency of the first qubit and the frequency of the second qubit through a first avoided crossing in a leakage channel; during a second stage, driving detuning between the frequency of the first qubit and the frequency of the second qubit through a second avoided crossing in a swap channel; during a third stage, evolving the first qubit and second qubit; during a fourth stage, implementing the second stage in reverse order; and during a fifth stage, implementing the first stage in reverse order.

Abstract: A method of resizing a block storage volume for a virtual machine includes executing the virtual machine and attaching a virtual storage device to the virtual machine. The virtual storage device exposes the block storage volume on memory hardware to the virtual machine. The block storage volume includes a first storage capacity. The method also includes mounting the block storage volume to the virtual machine and resizing the block storage volume while the virtual machine continues to execute. The block storage volume is resized without attaching an additional virtual storage device to the virtual machine or mounting an additional block storage volume to the virtual machine.

Abstract: A computer-implemented method includes receiving, by a computing device, a particular textual description of a scene. The method also includes applying a neural network for text-to-image generation to generate an output image rendition of the scene, the neural network having been trained to cause two image renditions associated with a same textual description to attract each other and two image renditions associated with different textual descriptions to repel each other based on mutual information between a plurality of corresponding pairs, wherein the plurality of corresponding pairs comprise an image-to-image pair and a text-to-image pair. The method further includes predicting the output image rendition of the scene.

Abstract: An example method includes receiving, by a computing system, an identification of a recipient of an electronic message being composed from a message account associated with a user; predicting, by the computing system and based on text contained in previous electronic messages sent from the message account, text for a body of the electronic message; and outputting, for display, the predicted text for optional selection and insertion into the body of the electronic message.

Abstract: A system and method of displaying, on a touchscreen display of a mobile device, a virtual game controller having at least one virtual button configured to control multiple actions of an avatar representing a player within a video game executed by a remote gaming server. The player's interaction with the at least one virtual button is received through the touchscreen display and used to generate a control signal based on the received player interaction, the control signal being interpretable by the video game as an instruction to perform multiple actions in the video game. The control signal is transmitted from the mobile device for receipt by the remote gaming server executing the video game.

Abstract: A computing device can include a cover, a texture haptics layer adjacent to the cover, a display layer adjacent to the texture haptics layer, an impact haptics layer adjacent to the display layer, a controller, and a housing enclosing the controller and supporting the cover, the texture haptics layer, the display layer, and the impact haptics layer. The controller can be configured to activate the texture haptics layer in response to an object moving along the cover, control an image presented by the display layer, and activate the impact haptics layer in response to the object contacting the cover.

Abstract: A first version of an electronic document is provided via a first region of a graphical user interface (GUI) for presentation to a first user of a collaborative document platform. A second region of the GUI includes an edit history GUI element that enables one or more users to access each edit made to the electronic document. A user selection of a portion of the edit history GUI element is detected. The user-selected portion of the edit history GUI element corresponds to edits made to the electronic document by a second user. A determination is made of a state of the electronic document following the edits made by the second user. The first region of the GUI is updated to include a rendering of a second version of the electronic document. The second version is rendered in view of the determined document state.

Abstract: Implementations are directed to determining, based on a submitted query that is a compound query, that a set of multiple sub-queries are collectively an appropriate interpretation of the compound query. Those implementations are further directed to providing, in response to such a determination, a corresponding command for each of the sub-queries of the determined set. Each of the commands is to a corresponding agent (of one or more agents), and causes the agent to generate and provide corresponding responsive content. Those implementations are further directed to causing content to be rendered in response to the submitted query, where the content is based on the corresponding responsive content received in response to the commands.

Abstract: The present disclosure provides a metal structure, or antenna radiator, for an accessory that is configured to function as both an antenna and an input control. The device may be a wearable device with a first surface shaped to be in contact with the human body and a second surface shaped to be exposed when the device is being worn. The antenna radiator may be located internally within the housing of the device and coupled to a printed circuit board (“PCB”). The PCB may be located between the antenna radiator and the surface of the housing such that the PCB may shield the antenna radiator from antenna loss due to the human body. An electric field may extend between the PCB and the antenna radiator. The fringe electric field may improve cross-head and cross-body connectivity between the accessory and another device.

Abstract: A method in a computing device includes generating a set of data packets comprising a first referrer data associated with a first resource rendered by the computing device when executing a first application and a second referrer data associated with a second resource rendered by the computing device prior to rendering the first resource. The first resource provides a first content item associated with a second application. The method also includes transmitting the set of data packets to an application server, receiving, from the application server, a deeplink, or a referrer tag, that was generated based on the first referrer data and the second referrer data, and retrieving, by the second application and using the deeplink or the referrer tag, content associated with the first referrer data and the second referrer data for display at the computing device within a content interface of the second application.

Abstract: Techniques are described in which a view control layer is optically coupled to a waveguide having one or more diffraction structures. The view control layer selectively prevents light rays from interfacing with at least one diffraction structure of the waveguide based on a respective angle of incidence at which the light rays encounter the view control layer. In various embodiments, the view control layer may include a light control film, a polarization stack, and/or a dielectric stack angular filter.

Abstract: A backplane operative to drive an array of emissive pixel elements forming a part of an automotive head lamp assembly is disclosed. Each pixel element comprises a memory cell operative to pulse width modulate a current mirror pixel drive circuit configured to drive an emissive element. The array of emissive pixel elements is divided into a plurality of interdigitated rows or columns serviced by independent row drivers or independent column drivers that may be driven by data selected to randomize the order in which the data on adjacent pixels of the same row are written, thereby effectively substantially reducing the visibility of any residual structures that may be present in the data driving the pixels of adjacent columns.

Abstract: Systems and methods for generating min-increment counting bloom filters to determine count and frequency of device identifiers and attributes in a networking environment are disclosed. The system can maintain a set of data records including device identifiers and attributes associated with device in a network. The system can generate a vector comprising coordinates corresponding to counter registers. The system can identify hash functions to update a counting bloom filter. The system can hash the data records to extract index values pointing to a set of counter registers. The system can increment the positions in the min-increment counting bloom filter corresponding to the minimum values of the counter registers. The system can obtain an aggregated public key comprising a public key. The system can encrypt the counter registers using the aggregated shared key to generate an encrypted vector. The system can transmit the encrypted vector to a networked worker computing device.

Abstract: Aspects of the technology utilize an integrated computing device that supports a shared, communal interaction while still enabling a personalized experience. This includes a unified desktop experience based on presence, a shared family or common space as part of an ambient visual display, and seamless switching involving user interface elements when the display device is rotatable. A presence sensor detects whether someone is at or near the device. When at least one person is identified, specific information is surfaced on a display based on the current context, enabling the person(s) to interact with the device in selected ways. The device evaluates a current presentation state of the display and selects content items to present. The presentation state can include content currently being displayed, other context signals, or any other information associated with the state of the device, such as whether the device is locked or logged-in, for example.

Abstract: A system is described that determines, based on information associated with a user of a computing device, an event for initiating an interaction between the user and an assistant executing at the computing device. The system selects, based on the event and from a plurality of actions performed by the assistant, at least one action associated with the event. The system determines, based on the at least one action, whether to output a notification of the event which includes an indication of the event and a request to perform the at least one action associated with the event. Responsive to determining to output the notification of the event, the system sends, to the assistant, the notification of the event for output during the interaction between the user and the assistant.

Abstract: A lighting stage includes a plurality of lights that project alternating spherical color gradient illumination patterns onto an object or human performer at a predetermined frequency. The lighting stage also includes a plurality of cameras that capture images of an object or human performer corresponding to the alternating spherical color gradient illumination patterns. The lighting stage also includes a plurality of depth sensors that capture depth maps of the object or human performer at the predetermined frequency. The lighting stage also includes (or is associated with) one or more processors that implement a machine learning algorithm to produce a three-dimensional (3D) model of the object or human performer. The 3D model includes relighting parameters used to relight the 3D model under different lighting conditions.

Abstract: A method for estimating an embedding capacity includes receiving, at a deterministic reference encoder, a reference audio signal, and determining a reference embedding corresponding to the reference audio signal, the reference embedding having a corresponding embedding dimensionality. The method also includes measuring a first reconstruction loss as a function of the corresponding embedding dimensionality of the reference embedding and obtaining a variational embedding from a variational posterior. The variational embedding has a corresponding embedding dimensionality and a specified capacity. The method also includes measuring a second reconstruction loss as a function of the corresponding embedding dimensionality of the variational embedding and estimating a capacity of the reference embedding by comparing the first measured reconstruction loss for the reference embedding relative to the second measured reconstruction loss for the variational embedding having the specified capacity.

Abstract: Methods and apparatus related to determining a triggering event of a user, selecting media relevant to the triggering event, and providing the selected media to the user. Some implementations are directed to methods and apparatus for determining a past event of the user that is indicative of past interaction of the user with one or more past entities and the triggering event may be determined to be associated with the past event. The media selected to provide to the user may contain media that includes the one or more past entities associated with the past event and the media may be provided to the user in response to the triggering event.