-
公开(公告)号：US10776716B2
公开(公告)日：2020-09-15
申请号：US15621753
申请日：2017-06-13
Applicant: Microsoft Technology Licensing, LLC
Inventor： Yu Liu , Jianshu Chen , Li Deng
IPC: G06N20/00 , G10L15/06 , G10L15/197 , G06N7/00
Abstract: In classification tasks applicable to data that exhibit sequential output statistics, a classifier may be trained in an unsupervised manner based on a sequence of input samples and an unaligned sequence of output labels, using a cost function that measures the negative cross-entropy of an N-gram joint probability distribution derived from the sequence of output labels with respect to an expected N-gram frequency in a second sequence of output labels predicted by the classifier. In some embodiments, a primal-dual reformulation of the cost function is employed to facilitate optimization.
-
公开(公告)号：US20190130360A1
公开(公告)日：2019-05-02
申请号：US15799490
申请日：2017-10-31
Applicant: Microsoft Technology Licensing, LLC
Inventor： Ke Wu , Yi Zhang , Hong Yao , Onkar A. Dalal , Yu Liu
Abstract: During operation, a system obtains member features associated with a member of a network, wherein the set of member features include a job-seeking status of the member. Next, the system analyzes the member features to predict an interest of the member in career services associated. The system then uses the predicted interest to output a recommendation of the career services to the member.
-
公开(公告)号：US11599746B2
公开(公告)日：2023-03-07
申请号：US16916706
申请日：2020-06-30
Applicant: Microsoft Technology Licensing, LLC
Inventor： Jilei Yang , Yu Liu , Parvez Ahammad , Fangfang Tan
Abstract: Techniques for detecting label shift and adjusting training data of predictive models in response are provided. In an embodiment, a first machine-learned model is used to generate a predicted label for each of multiple scoring instances. The first machine-learned model is trained using one or more machine learning techniques based on a plurality of training instances, each of which includes an observed label. In response to detecting a shift in observed labels, for each segment of one or more segments in multiple segments, a portion of training data that corresponds to the segment is identified. For each training instance in a subset of the portion of training data, the training instance is adjusted. The adjusted training instance is added to a final set of training data. The machine learning technique(s) are used to train a second machine-learned model based on the final set of training data.
-
公开(公告)号：US20210119125A1
公开(公告)日：2021-04-22
申请号：US17135632
申请日：2020-12-28
Applicant: Microsoft Technology Licensing, LLC
Inventor： Peter Krogstrup Jeppesen , Yu Liu , Alessandra Luchini
Abstract: A first aspect provides a topological quantum computing device comprising a network of semiconductor-superconductor nanowires, each nanowire comprising a length of semiconductor formed over a substrate and a coating of superconductor formed over at least part of the semiconductor; wherein at least some of the nanowires further comprise a coating of ferromagnetic insulator disposed over at least part of the semiconductor. A second aspect provides a method of fabricating a quantum or spintronic device comprising a heterostructure of semiconductor and ferromagnetic insulator, by: forming a portion of the semiconductor over a substrate in a first vacuum chamber, and growing a coating of the ferromagnetic insulator on the semiconductor by epitaxy in a second vacuum chamber connected to the first vacuum chamber by a vacuum tunnel, wherein the semiconductor comprises InAs and the ferromagnetic insulator comprises EuS.
-
公开(公告)号：US20210406598A1
公开(公告)日：2021-12-30
申请号：US16916706
申请日：2020-06-30
Applicant: Microsoft Technology Licensing, LLC
Inventor： Jilei Yang , Yu Liu , Parvez Ahammad , Fangfang Tan
Abstract: Techniques for detecting label shift and adjusting training data of predictive models in response are provided. In an embodiment, a first machine-learned model is used to generate a predicted label for each of multiple scoring instances. The first machine-learned model is trained using one or more machine learning techniques based on a plurality of training instances, each of which includes an observed label. In response to detecting a shift in observed labels, for each segment of one or more segments in multiple segments, a portion of training data that corresponds to the segment is identified. For each training instance in a subset of the portion of training data, the training instance is adjusted. The adjusted training instance is added to a final set of training data. The machine learning technique(s) are used to train a second machine-learned model based on the final set of training data.
-
Patent Agency Ranking
公开(公告)号：US20210357458A1
公开(公告)日：2021-11-18
申请号：US16877147
申请日：2020-05-18
Applicant: Microsoft Technology Licensing, LLC
Inventor： Jingwei Wu , Caleb Johnson , Yu Liu , Dianxia Yang , George James Pearman , Funing Xu
IPC: G06F16/9038 , G06Q10/06 , G06Q10/10
Abstract: Techniques are provided for automatically generating a real-time notification associated with a job posting. In one technique, in response to receiving a job posting, a plurality of saved searches that includes a first set of criteria and a second set of criteria is identified. For each saved search, a job posting is added to a first subset of saved searches if the job posting satisfies the first set of criteria. For each saved search of the first subset, a saved search is added to a second subset of saved searches if the job posting satisfies the second set of criteria. For each saved search of the second subset, a notification associated with the job posting is automatically sent to a computing device of a user associated with each saved search of the second subset.
-
公开(公告)号：US20200042946A1
公开(公告)日：2020-02-06
申请号：US16050559
申请日：2018-07-31
Applicant: Microsoft Technology Licensing, LLC
Inventor： Xiang Li , Yu Liu , Yunjing Zhang , Rishi Jobanputra
Abstract: The disclosed embodiments provide a system for inferring successful hires in an online system. During operation, the system samples feedback from users of an online system to generate labels representing hiring outcomes from requests for proposal (RFPs) submitted in the online system. Next, the system inputs the labels with features representing interaction associated with the RFPs as training data for a machine learning model. The system then applies one or more rules derived from the machine learning model to additional features for an additional RFP to infer a hiring outcome for the additional RFP. Finally, the system stores the inferred hiring outcome in association with the additional RFP.
-
公开(公告)号：US20190130464A1
公开(公告)日：2019-05-02
申请号：US15799496
申请日：2017-10-31
Applicant: Microsoft Technology Licensing, LLC
Inventor： Hong Yao , Yi Zhang , Yu Liu , Onkar A. Dalal , Thogori C. Karago , Ajita Thomas
Abstract: The disclosed embodiments provide a system for processing data. During operation, the system identifies a set of providers that meet a set of requirements in a request for proposal (RFP) from a consumer based on an overall compatibility between the set of requirements and the set of providers. Next, the system generates a ranking of the set of providers based on the compatibility and a network distance between the consumer and the set of providers in a social network. The system then selects one or more providers from the ranking as matches for the RFP. Finally, the system transmits the RFP to the one or more providers.
-
公开(公告)号：US11424409B2
公开(公告)日：2022-08-23
申请号：US17135632
申请日：2020-12-28
Applicant: Microsoft Technology Licensing, LLC
Inventor： Peter Krogstrup Jeppesen , Yu Liu , Alessandra Luchini
Abstract: A first aspect provides a topological quantum computing device comprising a network of semiconductor-superconductor nanowires, each nanowire comprising a length of semiconductor formed over a substrate and a coating of superconductor formed over at least part of the semiconductor; wherein at least some of the nanowires further comprise a coating of ferromagnetic insulator disposed over at least part of the semiconductor. A second aspect provides a method of fabricating a quantum or spintronic device comprising a heterostructure of semiconductor and ferromagnetic insulator, by: forming a portion of the semiconductor over a substrate in a first vacuum chamber, and growing a coating of the ferromagnetic insulator on the semiconductor by epitaxy in a second vacuum chamber connected to the first vacuum chamber by a vacuum tunnel, wherein the semiconductor comprises InAs and the ferromagnetic insulator comprises EuS.
-
公开(公告)号：US11182432B2
公开(公告)日：2021-11-23
申请号：US16457222
申请日：2019-06-28
Applicant: Microsoft Technology Licensing, LLC
Inventor： Jun Shi , Huiji Gao , Ying Xiong , Michaeel M. Kazi , Yu Gan , Yu Liu , Xiaowei Liu , Gonzalo Jorge Aniano Porcile , Bo Long , Abhimanyu Lad , Liang Zhang
IPC: G06F17/00 , G06F16/9032 , G06N20/00 , G06F16/2457 , G06F40/30
Abstract: The disclosed embodiments provide a system for performing a natural language search. During operation, the system applies a first machine learning model to a natural language query to predict one or more search intentions associated with the natural language query. Next, the system applies a second machine learning model to the natural language query to produce one or more search parameters associated with a first intention in the search intention(s), wherein the search parameter(s) include a field and a value of the field. The system then performs a first search of a first vertical associated with the first intention using the search parameter(s). Finally, the system generates a ranking containing a first set of search results from the first search of the first vertical and outputs at least a portion of the ranking in a response to the natural language query.
-
-
-
-
-
-
-
-
-
Search Results
13
records
(took 0.019 seconds)
