Microsoft Technology Licensing, LLC

Other

David J Stein

Joel Daniel Young

Harold M Stein

Krishnaram Kenthapadi

Shaunak Chatterjee

Weidong Zhang

Bee-Chung Chen

Bo Zhao

Parul Jain

Eric Huang

Grace W Tang

Maneesh Varshney

Xu Miao

Sumit Rangwala

Boyi Chen

Wei Zhuang

Fedor Vladimirovich Borisyuk

Amit Chandak

Paul Taylor Ogilvie

Wenxuan Gao

Jian Qiao

Da Teng

Songxiang Gu

Chang-Ming Tsai

Lance M Wall

Varnit Agnihotri

Divya Gadde

Ruoyang Wang

Amit Yadav

Yangchun Luo

Min Shen

Ke Wu

Priyanka Gariba

The disclosed embodiments provide a system for processing data. During operation, the system obtains a feature configuration for a feature. Next, the system obtains, from the feature configuration, an anchor containing metadata for accessing the feature in an environment. The system then uses one or more attributes of the anchor to retrieve one or more feature values of the feature from the environment. Finally, the system provides the one or more feature values for use with one or more machine-learning models.

Framework for managing features across environments

The disclosed embodiments provide a system for processing data. During operation, the system organizes fact data to be aggregated into sliding time window features and observation data associated with the fact data into a set of partitions based on a join key. Next, the system sorts the fact data and the observation data within the set of partitions by the join key and timestamps associated with the fact data and the observation data. For each observation record in the observation data, the system aggregates fact records in the sorted fact data that share a value of the join key with the observation record and that fall within a first time window associated with the observation record to produce a sliding time window feature. The system then stores the sliding time window feature in association with the observation record.

Efficient aggregation of sliding time window features

The disclosed embodiments provide a system for processing data. During operation, the system obtains a hierarchical representation containing a set of namespaces of a set of features shared by a set of statistical models. Next, the system uses the hierarchical representation to obtain, from one or more execution environments, a subset of the features for use in calculating the derived feature. The system then applies a formula from the hierarchical representation to the subset of the features to produce the derived feature. Finally, the system provides the derived feature for use by one or more of the statistical models.

Common feature protocol for collaborative machine learning

Methods and systems for generating queries from a behavioral index model are described. According to various embodiments, a system receives a query input including one or more search terms from a client device associated with a member profile. The system generates a query from the one or more search terms and one or more profile terms. The system causes one or more query machines to search one or more inverted indices using the query and receives one or more result of documents distributed across the inverted indices. The system generates an ordered list of the one or more result sets and causes presentation of the ordered list at a client device associated with the member profile.

Machine learned query generation on inverted indices

Various embodiments of a rod for cleaning and/or polishing embodiments of a ferrule in a fiber optic connector are disclosed. In a first embodiment, the rod is formed of an ultra high molecular weight polyethylene hydrophobic material, preferably absorbent. Each end of the rod has a hole that is a tight fit to the diameter of a male-type ferrule and is at least as deep as the length of the ferrule. The rod material is strong and pliable, so that after it is slipped over the male-type ferrule the rod can be rotated for cleaning the sides of the ferrule while being withdrawn from its snug fit with the ferrule. The holes may have polishing inserts formed therein. Alternatively, the rod has a male protrusion on one end. The male end inserts into the recess formed by the connector adaptor when using a female-type ferrule. The rod is then rotated to effect cleaning and/or polishing of the end face of the optical fiber together with cleaning of the insides surfaces of the side walls of the connector adaptor.

Ferrule cleaning rod and method of use

A mechanical connector that normally interfaces two fiber optic cables together has a fiber optic cable inserted in a receptacle on one side of the connector. A device for cleaning and/or polishing the end surface of an optical fiber strand within the cable is disposed in a second receptacle on the opposite side of the connector. The device includes an insert located within the receptacle in the connector, the insert having a hole drilled therethrough that is eccentric to the transmission axis of the optical fiber strand. A cleaning rod may be inserted in the eccentric hole in the insert such that an end surface of the cleaning rod contacts the end surface of the optical fiber strand therein. The cleaning rod is rotated to effectuate cleaning of the end surface of the fiber optic cable and corresponding optical fiber strand. In the alternative, the polishing rod may be inserted in the eccentric hole in the insert and rotated so as to effectuate polishing of the end surface of the optical fiber strand.

Device for cleaning and polishing an optical fiber

Growing community of inventors

Mountain View, CA, United States of America

David J Stein