ASYNCHRONOUS PREDICTION IN MACHINE LEARNING MODEL SERVING

    公开(公告)号:US20240095109A1

    公开(公告)日:2024-03-21

    申请号:US17945960

    申请日:2022-09-15

    申请人: Twitter, Inc.

    发明人: Lawrence Lam Di Zhao

    IPC分类号: G06F9/54 G06N20/00

    CPC分类号: G06F9/547 G06N20/00

    摘要: Example computer-implemented methods, media, and systems for serving machine learning (ML) models using an asynchronous input/output (I/O) mechanism are disclosed. One example method includes receiving a first request for running a ML model to provide a first prediction. A first green thread is generated responsive to the first request and executed on an operating system (OS) thread to send a first asynchronous remote procedure call (RPC) to a multiple producer single consumer (MPSC) channel. A second request for running the ML model to provide a second prediction is received. A second green thread is generated responsive to the second request and executed on the OS thread to send a second RPC to the MPSC channel. The first and the second asynchronous RPCs are scheduled using a first and a second blocking threads respectively, which are used by the ML model to generate the first prediction and the second prediction.

    RANKING MESSAGES OF A CONVERSATION GRAPH AND DISPLAYING SOCIAL CONTEXT OBJECTS

    公开(公告)号:US20230306068A1

    公开(公告)日:2023-09-28

    申请号:US17805970

    申请日:2022-06-08

    申请人: Twitter, Inc.

    摘要: According to an aspect, a method includes receiving, over a network, a conversation view request to retrieve messages from a conversation graph stored on a messaging platform from a first user account, generating a ranked list of messages for at least a portion of the messages from the conversation graph, identifying, from the ranked list of messages, a message authored by a second user account having an indirect relationship to the first user account in an engagement graph, selecting a social context object for the message based on a type of the indirect relationship, and transmitting, over the network, information to display the message and the social context object, the social context object providing information about the type of the indirect relationship.

    Prioritizing messages within a message network

    公开(公告)号:US11606323B2

    公开(公告)日:2023-03-14

    申请号:US17020624

    申请日:2020-09-14

    申请人: Twitter, Inc.

    摘要: A system and a method are disclosed for recommending electronic messages in a message sharing system. Users can post messages to the message sharing system. These messages from posting users are received by the system and sent to receiving users that have subscribed to the posting users. The receiving users interact with the messages in various ways, such as by sharing the messages with other users. Interaction information is received for each of the electronic messages. The interaction information includes an indication of the number of interactions with the electronic message by receiving users. A score is determined for each electronic message based on the interaction information. Electronic messages are selected for being recommended to a user or a group of users based on the scores. The recommendations are then sent to the users, enabling users to better focus their attention on messages that are likely to be interesting.

    Determining whether a user in a social network is an authority on a topic

    公开(公告)号:US11567947B1

    公开(公告)日:2023-01-31

    申请号:US17214715

    申请日:2021-03-26

    申请人: Twitter, Inc.

    发明人: Alek Kolcz

    摘要: A method involving obtaining a first plurality of topic groups (TGs), each having a membership of accounts, identifying a first plurality of accounts as authorities for an expertise topic, obtaining a second plurality of TGs with a number of accounts as members, wherein the first plurality of TGs comprises the second plurality of TGs, identifying a first frequent account which is a member in at least one of the second plurality of TGs, adding the first frequent account to the authorities of the expertise topic to obtain a second plurality of accounts as the authorities of the expertise topic, determining a third plurality of TGs in which a second number of accounts from the second plurality of accounts are members, determining that another frequent account is a member in one of the third plurality of TGs, and obtaining a ranking of accounts that are an authority on the expertise topic.

    Messaging-enabled unmanned aerial vehicle

    公开(公告)号:US11556123B2

    公开(公告)日:2023-01-17

    申请号:US16915056

    申请日:2020-06-29

    申请人: Twitter, Inc.

    IPC分类号: G06F17/00 G05D1/00 H04L51/52

    摘要: An unmanned aerial vehicle (UAV) carries a camera, sends data from the camera, and receives commands. The UAV is connected to a messaging platform. Pictures or video clips received from the UAV are selected and placed in messages broadcast by an account associated with the UAV. Video footage from the camera is live-streamed in a card-type message. Account holders of the messaging platform may control the UAV with commands embedded in messages and directed towards an account associated with the UAV. Controllable elements of the UAV include UAV location, camera orientation, camera subject, UAV-mounted lighting, a UAV-mounted display, a UAV-mounted projector, UAV-mounted speakers, and a detachable payload. UAV control may be determined through democratic means. Some UAV functionality may be triggered through aggregated engagements on the messaging platform. The UAV may include a display screen and/or a microphone to provide for telepresence or interview functionality.

    User-defined message timelines
    6.
    发明授权

    公开(公告)号:US11552919B1

    公开(公告)日:2023-01-10

    申请号:US16147552

    申请日:2018-09-28

    申请人: Twitter, Inc.

    摘要: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for creating and managing custom timelines. One of the methods includes receiving, from a user using a user account of a social messaging platform, input to a user interface displaying a message timeline of a social messaging platform, the user input requesting that the platform save a first message from the message timeline and, in response, saving the message to a custom timeline of the user account, wherein saving the message to the custom timeline includes associating with the custom timeline an identifier that identifies the first message; monitoring engagements by one or more other accounts with the first message with respect to one or more notification criteria; and in response to determining that the monitored engagements satisfy a trigger condition, providing a notification to the user account associated with the first message through the custom timeline.

    AGGREGATE FEATURES FOR MACHINE LEARNING

    公开(公告)号:US20220382564A1

    公开(公告)日:2022-12-01

    申请号:US17819134

    申请日:2022-08-11

    申请人: Twitter, Inc.

    摘要: An example system includes a memory store of aggregate definitions. Each aggregate definition specifies a key value, a feature, a half-life value, and an aggregate operation metric to apply to a cross of the feature and the half-life value to generate aggregate metrics. The system also includes an aggregation engine that generates aggregate feature records from the input source based on the aggregate definitions and stores the aggregate feature records. An aggregate feature record includes an aggregate of the metric for the feature decayed over time using the half-life. The system also includes a query service that identifies, using the aggregate definitions, responsive aggregate feature records that satisfy parameters of a received request, applies the half-life to the responsive feature records, and provides the responsive feature records to a requester, the requester using the responsive feature records as input for a neural network.

    Hard and soft ranking messages of conversation graphs in a messaging platform

    公开(公告)号:US11516155B1

    公开(公告)日:2022-11-29

    申请号:US16723987

    申请日:2019-12-20

    申请人: Twitter, Inc.

    摘要: A method for ranking messages of conversation graphs in a messaging platform includes classifying messages of a conversation graph into a plurality of sections based on content quality of the messages, where the plurality of sections include a first section and a second section. The first section has messages from the conversation graph determined as higher quality than messages of the second section. The method includes determining a plurality of predictive outcomes for each of a plurality of messages of the conversation graph using predictive models, ranking the messages of the first section based on the predictive outcomes for the messages of the first section, and transmitting at least a subset of the messages of the first section to be rendered on a client application according to the rank.