Microsoft Technology Licensing, LLC

Yiming Ma

Bee-Chung Chen

Bo Long

Liang Zhang

Deepak Agarwal

Wei Guang Lu

Ajith Muralidharan

Yitong Zhou

Yi Zhen

Xuhong Zhang

Florian Raudies

Mingzhou Zhou

Josh Fleming

Menglin L Brown

Alex Shelkovnykov

XianXing Zhang

Jun Jia

Sheng Wu

The disclosed embodiments provide a system for processing data. During operation, the system applies a first set of hash functions to a first entity identifier (ID) for a first entity to generate a first set of hash values. Next, the system produces a first set of intermediate vectors from the first set of hash values and a first set of lookup tables by matching each hash value in the first set of hash values to an entry in a corresponding lookup table in the first set of lookup tables. The system then performs an element-wise aggregation of the first set of intermediate vectors to produce a first embedding. Finally, the system outputs the first embedding for use by a machine learning model.

Compact entity identifier embeddings

The disclosed embodiments provide a system for processing data. During operation, the system obtains a training dataset containing a first set of records associated with a first set of identifier (ID) values and an evaluation dataset containing a second set of records associated with a second set of ID values. Next, the system selects a random subset of ID values from the second set of ID values. The system then generates a sampled evaluation dataset comprising a first subset of records associated with the random subset of ID values in the second set of records. The system also generates a sampled training dataset comprising a second subset of records associated with the random subset of ID values in the first set of records. Finally, the system outputs the sampled training dataset and the sampled evaluation dataset for use in training and evaluating a machine learning model.

Data sampling for model exploration utilizing a plurality of machine learning models

In an example embodiment, a warm-start training solution is used to dramatically reduce the computational resources needed to train when retraining a generalized additive mixed-effect (GAME) model. The problem of retraining time is particularly applicable to GAME models, since these models take much longer to train as the data grows. In the past, the strategy to reduce computational resources during retraining was to use less training data, but this affects the model quality, especially for GAME models, which rely on fine-grained sub-models at, for example, member or item levels. The present solution addresses the computational resources issues without sacrificing GAME model accuracy.

Warm start generalized additive mixed-effect (game) framework

Disclosed are systems, methods, and non-transitory computer-readable media for automated profile image generation based on scheduled video conferences. A profile image generation system generates, based on image data captured during a first video conference, a first facial feature data set for a first identified face identified from the image data. The first facial feature data set includes numeric values representing the first identified face. The profile image generation system calculates, based on the first facial feature data set and historic facial feature data sets generated from image data captured during previous video conferences, a first value indicating a likelihood that the first identified face is a first meeting participant that participated in the first video conference. The profile image generation system determines that the first value meets or exceeds a threshold value, and in response, determines that the first identified face is the first meeting participant.

Automated profile image generation based on scheduled video conferences

The disclosed embodiments provide a system for processing data. During operation, the system obtains a global version of a statistical model and a user-specific version of the statistical model for a user. Next, the system applies the global version to member features of the user and job features of a set of jobs to generate a first ranking of the jobs for the user. The system then applies the user-specific version to the member features and the job features for a highest-ranked subset of jobs in the first ranking to generate a second ranking of the jobs for the user. Finally, the system outputs at least a portion of the second ranking as a set of job recommendations.

Scalable response prediction using personalized recommendation models

Growing community of inventors

Menlo Park, CA, United States of America

Yiming Ma