Microsoft Technology Licensing, LLC

Huiji Gao

Bo Long

Jianling Zhong

Haichao Wei

Weiwei Guo

Lin Guo

Jun Shi

Haishan Liu

Michaeel M Kazi

Yuwei Qiu

Yu Gan

Gonzalo Jorge Aniano Porcile

Liang Zhang

Kun Liu

Bee-Chung Chen

Yuan Gao

Abhimanyu Lad

Yu Liu

Qin Iris Wang

Liqin Xu

Jan Schellenberger

Chengming Jiang

Zhong Yi Wan

Ying Xiong

Zhiling Wan

Chih-Hui Wang

Xiaowei Liu

Lili Zhou

Qiang Xiao

In an example, a particular type of deep learning model is used in the global model of the GDMix model: a Factorization Machine. A Factorization Machine combines a Support Vector Machine (SVM) and Matrix Factorizations. It has the advantage of modeling data with huge sparsity well, while maintaining a linear time complexity. A modification may be further made to the Factorization Machine by introducing Lnorm reduction. This acts to divide calculations made by the Factorization Machine into a portion that can be precomputed and a portion that cannot be precomputed. The portion that can be precomputed is then precomputed in an offline manner. As such, when the model is operated in an online manner, the Factorization Machine only needs to compute the portion that cannot be precomputed, reducing the number of operations that need to performed at runtime and greatly improving processing speed over prior machine learned models.

Factorization machine with L-2 norm reduction for machine learned models

Embodiments of the described technologies are capable of reading a text sequence that include at least one word; extracting model input data from the text sequence, where the model input data includes, for each word of the text sequence, segment data and non-segment data; using a first machine learning model and at least one second machine learning model, generating, for each word of the text sequence, a multi-level feature set; outputting, by a third machine learning model, in response to input to the third machine learning model of the multi-level feature set, a tagged version of the text sequence; executing a search based at least in part on the tagged version of the text sequence.

Neural tagger with deep multi-level model

Embodiments of the disclosed technologies use machine learning to produce thread level classification data and case level classification data.

Domain-adapted sentiment prediction for long or unbalanced text threads

Computer-implemented techniques for learning composite machine learned models are disclosed. Benefits to implementors of the disclosed techniques include allowing non-machine learning experts to use the techniques for learning a composite machine learned model based on a learning dataset, reducing or eliminating the explorative trial and error process of manually tuning architectural parameters and hyperparameters, and reducing the computing resource requirements and model learning time for learning composite machine learned models. The techniques improve the operation of distributed learning computing systems by reducing or eliminating straggler effects and by reducing or minimizing synchronization latency when executing a composite model search algorithm for learning a composite machine learned model.

Distributed learning of composite machine learning models

Machine learning based method for generating personalized query suggestions is described. Different users may have different search intent even when they are inputting the same search query. The technical problem of personalizing search query suggestions produced by a machine learning model is addressed by extending the sequence to sequence machine learning model framework to be able to take into consideration additional, personalized features of the user, such as, e.g., profile industry, language, geographic location, etc. This methodology includes an offline model training framework as well as an online serving framework.

Generating personalized query suggestions

A neural related query generation approach in a search system uses a neural encoder that reads through a source query to build a query intent vector. The approach then processes the query intent vector through a neural decoder to emit a related query. By doing so, the approach gathers information from the entire source query before generating the related query. As a result, the neural encoder-decoder approach captures long-range dependencies in the source query such as, for example, structural ordering of query keywords. The approach can be used to generate related queries for long-tail source queries, including long-tail source queries never before or not recently submitted to the search system.

Neural related search query generation

Techniques for extracting features of entities and targets that can be applied in a set of applications, such as entity selection prediction, audience expansion, feed relevance, and job recommendation. In one technique, entity interaction data is stored that indicates, for each of multiple entities, one or more targets that are associated with items with which the entity interacted. Token association data is stored that indicates, for each of multiple tokens, one or more targets that are associated with the token. Then, using one or more machine learning techniques, entity embeddings and target embeddings are generated based on the entity interaction data and the token association data. Later, a request for content is received from a particular entity. Based on at least one entity embedding, a content item for the particular entity is identified. The content item is transferred over a computer network and presented to the particular entity.

Generating machine-learned entity embeddings based on online interactions and semantic context

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.

Vertical processing of natural language searches

In an embodiment, the disclosed technologies include extracting, from at least one search log, session data including at least three semantically related queries and corresponding timestamp data; using the session data, creating a training sequence that includes source query data, context query data, and target query data, the source query data having both a temporal relationship and a lexical relationship to the target query data and the context query data having a temporal relationship to the source query data; creating a learned model by, using a machine learning-based modeling process, learning a mapping of a semantic representation of the context query data and the source query data to a semantic representation of the target query data; in response to a new query, using the learned model to generate at least one recommended query that is semantically related to the new query.

Session-aware related search generation

Herein are techniques to use an artificial neural network to score the relevance of content items for a target and techniques to rank the content items based on their scores. In embodiments, a computer uses a plurality of expansion techniques to identify expanded targets for a content item. For each of the expanded targets, the computer provides inputs to an artificial neural network to generate a relevance score that indicates a relative suitability of the content item for that target. The computer ranks the expanded targets based on the relevance score generated for each of the expanded targets. Based on the ranking, the computer selects a subset of targets from the available expanded targets as the expanded targets for whom the content item is potentially most relevant. The computer stores an association between the content item and each target in the subset of expanded targets.

Deep neural networks for targeted content distribution

Machine learning techniques are described for generating recommendations using decision trees. A decision tree is generated based on training data that comprises multiple training instances, each of which comprises a feature value for each of multiple features and a label of a target variable. The multiple features correspond to attributes of multiple content delivery campaigns. Later, feature values of a content delivery campaign are received. The decision tree is traversed using the feature values to generate output. Based on the output, one or more recommendations are identified and the one or more recommendations are presented on a computing device.

Machine-learned recommender system for performance optimization of network-transferred electronic content items

A system and method for determining whether a computer system is experiencing a problem are provided. Multiple data sets are identified. Each data set includes multiple values indicating a set of attributes that relate to a computer system and that correspond to a different time period of multiple time periods, such as days or hours. A model is generated based on the multiple data sets. A particular data set is identified that includes a set of values that correspond to the set of attributes and a particular time period that is subsequent to each of the multiple time periods. The model is used to generate a predicted value based on the particular data set. An actual value that corresponds to the particular time period is identified. A difference between the actual value and the predicted value is calculated and indicates a likelihood that the computer system is experiencing a problem.

Growing community of inventors

San Jose, CA, United States of America

Huiji Gao