Abstract: Methods, systems, and apparatus are described for throttling a distributed processing system. In one aspect, a method includes identifying records being processed by a distributed processing system that performs agent processes, each of the records including a corresponding timestamp; determining, based on timestamps of the records that have been processed by a first agent process, a first agent progress; identifying a dependent agent process performed by the distributed processing system, wherein the dependent agent process processes only records that have been processed by the first agent process; determining, based on timestamps of records that have been processed by the dependent agent process, a dependent agent progress; and throttling performance of the first process based on the first agent progress and the dependent agent progress.

Abstract: Various embodiments of systems and methods for image matching and retrieval process are disclosed.

Abstract: Provided is a method of spatially referencing a plurality of images captured from a plurality of different locations within an indoor space by determining the location from which the plurality of images was captured. The method may include obtaining a plurality of distance-referenced panoramas of an indoor space. The distance-referenced panoramas may each include a plurality of distance-referenced images each captured from one position in the indoor space and at a different azimuth from the other distance-referenced images, a plurality of distance measurements, and orientation indicators each indicative of the azimuth of the corresponding one of the distance-referenced images. The method may further include determining the location of each of the distance-referenced panoramas based on the plurality of distance measurements and the orientation indicators and associating in memory the determined locations with the plurality of distance-referenced images captured from the determined location.

Abstract: Systems and techniques are provided for template exploration in a large-scale machine learning system. A method may include obtaining multiple base templates, each base template comprising multiple features. A template performance score may be obtained for each base template and a first base template may be selected from among the multiple base templates based on the template performance score of the first base template. Multiple cross-templates may be constructed by generating a cross-template of the selected first base template and each of the multiple base templates. Performance of a machine learning model may be tested based on each cross-template to generate a cross-template performance score for each of the cross-templates. A first cross-template may be selected from among the multiple cross-templates based on the cross-template performance score of the cross-template. Accordingly, the first cross-template may be added to the machine learning model.

Abstract: Various embodiments of systems and methods for image matching and retrieval process are disclosed.

Abstract: This specification describes a power distribution system comprising a first section that receives power from a first source. The power received from the first source is adjusted by a first rectifier unit coupled to a first power bus of the first section. The system also comprises a second section that is separate from the first section and that receives power from a second source. The power received from the second source is adjusted by a second rectifier unit coupled to a second power bus of the second distribution section. The system includes a swing rectifier connected to the first section and connected to the second section. The swing rectifier is configured to provide power to the first power bus and to the second power bus and to dynamically adjust the power capacity of the first section that is available to computing loads, and to dynamically adjust the power capacity of the second section that is available to computing loads.

Abstract: Enhanced Coulomb repulsion screening around light element nuclei is achieved by way of utilizing electromagnetic (EM) radiation to induce plasmon oscillations in target structures (e.g., nanoparticles) in a way that produces high density electron clouds in localized regions of the target structures, thereby generating charge density variations around light element atoms located in the localized regions. Each target structure includes an electrically conductive body including light elements (e.g., a metal hydride/deuteride/tritide) that is configured to undergo plasmon oscillations in response to the applied EM radiation. The induced oscillations causes free electrons to converge in the localized region, thereby producing transient high electron charge density levels that enhance Coulomb repulsion screening around light element (e.g., deuterium) atoms located in the localized regions.

Abstract: A system and method is provided whereby more data packets than are necessary may be transmitted in order to prompt a node of network to more quickly increase the number of packets that will be accepted by the node. In one aspect, a single packet of data is split into multiple packets. In another aspect, a single packet of data is padded so that multiple packets need to be sent.

Abstract: Systems and methods that automatically identify and extract or distinguish sub-conversations within a live chat session, are provided. In an aspect, a system includes a reception component configured to receive comments associated with a common topic, wherein the comments are displayed via a first graphical user interface. The system further includes an identification component configured to identify a subset of the comments associated with a sub-topic within the common topic, wherein in response to identification of the subset of comments, the subset of comments are distinguished among the comments as displayed via the first graphical user interface or another graphical user interface.

Abstract: A method for applying a style to an input image to generate a stylized image. The method includes maintaining data specifying respective parameter values for each image style in a set of image styles, receiving an input including an input image and data identifying an input style to be applied to the input image to generate a stylized image that is in the input style, determining, from the maintained data, parameter values for the input style, and generating the stylized image by processing the input image using a style transfer neural network that is configured to process the input image to generate the stylized image.

Abstract: An image sensor of an image capture device may capture an image. The captured image may be stored in a buffer of two or more previously-captured images. An oldest image of the two or more previously-captured images may be removed from the buffer. An aggregate image of the images in the buffer may be updated. This updating may involve subtracting a representation of the oldest image from the aggregate image, and adding a representation of the captured image to the aggregate image. A viewfinder of the image capture device may display a representation of the aggregate image.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for optimizing transmission of digital components. In one aspect a digital component distribution system performs operations including determining a minimum eligibility value required to maintain the eligibility of a given digital component based on a next highest eligibility value of one of the other digital components. Based on the next highest eligibility value, a baseline transmission value required to be submitted for distribution of the given digital component is determined. A normalized transmission value is determined based on the minimum eligibility value, and transmission of the given digital component to a client device presenting a particular electronic document is triggered based on the normalized transmission value.

Abstract: Implementations generally relate to providing recommended actions for photos. In some implementations, a method includes obtaining activity data associated with one or more actions of a performing user, wherein the activity data involves at least one first photo. The method further includes performing or suggesting one or more actions to one or more second photos based on one or more predetermined similarity criteria.

Abstract: A method includes: providing a first wafer including a first substrate, a first insulator layer on the first substrate, and a first dielectric layer on the first insulator layer; providing a second wafer including a second substrate, a second insulator layer on the second substrate, and a second dielectric layer on the second insulator layer; forming a first superconductor layer on the first dielectric layer; forming a second superconductor layer on the second dielectric layer; joining a surface of the first superconductor layer to a surface of the second superconductor layer to form a wafer stack; and forming a third superconductor layer on exposed first surface of the first dielectric layer.

Abstract: A user interface exposes an application feature of an advertising management application. A graduation trigger is identified, the graduation trigger providing a basis for exposing an additional application feature. The user interface is modified to expose the additional application feature based on the graduation trigger.

Abstract: In one example, the disclosure is directed to providing a multi-dimensional audio interface in which a plurality of different spatial locations is defined, wherein each of the plurality of different spatial locations is based on a common reference point and is uniquely identifiable by a computing device; associating, by the computing device, an information type with a first spatial location of the plurality of different spatial locations, the information type included in a plurality of different information types; responsive to determining that first information is of the information type, outputting, using an audio output device and simulating localization of sound at the first spatial location, a first sound that indicates the first information; receiving an indication of user input that selects the first information; and performing at least one operation corresponding to the first information selected based at least in part on the indication of user input.

Abstract: An internet-based content platform that includes subscribable channels is provided. The internet-based content platform provides channels to end users. The internet-based content platform allows a channel curator to identify the paid channel comprising a plurality of media items. The internet-based content platform allows the channel curator indicate a list of types of access to the plurality of media items, a list of subscription levels, where each subscription level is associated with a different subscription fee and where each subscription level is associated with a type of access from the list of types of access. The internet-based content platform associates the list of types of access and the list of subscription levels with the paid channel.

Abstract: A computer-implemented method can include identifying one or more candidate topics from a query. The method can generate, for each candidate topic, a candidate topic-answer pair that includes both the candidate topic and an answer to the query for the candidate topic. The method can obtain search results based on the query, wherein one or more of the search results references an annotated resource. For each candidate topic-answer pair, the method can determine a score for the candidate topic-answer pair for use in determining a response to the query, based on (i) an occurrence of the candidate topic in the annotations of the resources referenced by one or more of the search results, and (ii) an occurrence of the answer in annotations of the resources referenced by the one or more search results, or in the resources referenced by the one or more search results.

Abstract: A method includes identifying a plurality of items, each having a score and being sorted within a list, identifying one or more variation features, diversifying the list of the plurality of items by processing each of the plurality of items in order of the sorting, the processing for each of the plurality of items including selecting the item as a candidate item, determining one or more demotion criteria with respect to the candidate item, determining if one or more items of the plurality of items meet the demotion criteria with respect to the selected item, modifying the score for at least one of the plurality of items based on a demotion factor, rearranging the list according to the score of each of the plurality of items in response to the modifying and providing the list of the plurality of items for display to the user.