COMBINING MATH-PROGRAMMING AND REINFORCEMENT LEARNING FOR PROBLEMS WITH KNOWN TRANSITION DYNAMICS

    公开(公告)号:US20230041035A1

    公开(公告)日:2023-02-09

    申请号:US17751625

    申请日:2022-05-23

    IPC分类号: G06N3/04 G06F17/18

    摘要: A computer implemented method of improving parameters of a critic approximator module includes receiving, by a mixed integer program (MIP) actor, (i) a current state and (ii) a predicted performance of an environment from the critic approximator module. The MIP actor solves a mixed integer mathematical problem based on the received current state and the predicted performance of the environment. The MIP actor selects an action a and applies the action to the environment based on the solved mixed integer mathematical problem. A long-term reward is determined and compared to the predicted performance of the environment by the critic approximator module. The parameters of the critic approximator module are iteratively updated based on an error between the determined long-term reward and the predicted performance.

    Personalized low latency communication

    公开(公告)号:US11336596B2

    公开(公告)日:2022-05-17

    申请号:US13915547

    申请日:2013-06-11

    摘要: Embodiments relate to personalized low latency communications. A method may include receiving a description of content of a message, receiving recipient data corresponding to at least two possible recipients within a population of possible recipients, and selecting a relevant subpopulation of the population. The selecting may include, for each of the at least two possible recipients, ranking a strength of an indirect relationship between the description and the recipient data. The indirect relationship may be based on the description, the recipient data and at least one additional data source. The selecting may also include, for each of the at least two possible recipients, adding a possible recipient to the relevant subpopulation based on the ranking of the indirect relationship associated with the possible recipient. The method may further include initiating a two-way communication channel between a sender of the message and the relevant subpopulation.

    Joint embedding of corpus pairs for domain mapping

    公开(公告)号:US10657189B2

    公开(公告)日:2020-05-19

    申请号:US15240660

    申请日:2016-08-18

    摘要: Techniques for outside-in mapping for corpus pairs are provided. In one example, a computer-implemented method comprises: inputting first keywords associated with a first domain corpus; extracting a first keyword of the first keywords; inputting second keywords associated with a second domain corpus; generating an embedded representation of the first keyword via a trained model and generating an embedded representation of the second keywords via the trained model; and scoring a joint embedding affinity associated with a joint embedding. The scoring the joint embedding affinity comprises: transforming the embedded representation of the first keyword and the embedded representation of the second keywords via the trained model; determining an affinity value based on comparing the first keyword to the second keywords; and based on the affinity value, aggregating the joint embedding of the embedded representation of the first keyword and the embedded representation of the second keywords within the second domain corpus.

    Joint embedding of corpus pairs for domain mapping

    公开(公告)号:US10579940B2

    公开(公告)日:2020-03-03

    申请号:US15240649

    申请日:2016-08-18

    摘要: Techniques for outside-in mapping for corpus pairs are provided. In one example, a computer-implemented method comprises: inputting first keywords associated with a first domain corpus; extracting a first keyword of the first keywords; inputting second keywords associated with a second domain corpus; generating an embedded representation of the first keyword via a trained model and generating an embedded representation of the second keywords via the trained model; and scoring a joint embedding affinity associated with a joint embedding. The scoring the joint embedding affinity comprises: transforming the embedded representation of the first keyword and the embedded representation of the second keywords via the trained model; determining an affinity value based on comparing the first keyword to the second keywords; and based on the affinity value, aggregating the joint embedding of the embedded representation of the first keyword and the embedded representation of the second keywords within the second domain corpus.