公开(公告)号：US20210065007A1

公开(公告)日：2021-03-04

申请号：US16551615

申请日：2019-08-26

申请人： Microsoft Technology Licensing, LLC

发明人： Matteo INTERLANDI ,  Byung-Gon CHUN ,  Markus WEIMER ,  Gyeongin YU ,  Saeed AMIZADEH

IPC分类号： G06N3/08 ,  G06N20/00 ,  G06T1/20 ,  G06F16/901

摘要： Solutions for adapting machine learning (ML) models to neural networks (NNs) include receiving an ML pipeline comprising a plurality of operators; determining operator dependencies within the ML pipeline; determining recognized operators; for each of at least two recognized operators, selecting a corresponding NN module from a translation dictionary; and wiring the selected NN modules in accordance with the operator dependencies to generate a translated NN. Some examples determine a starting operator for translation, which is the earliest recognized operator having parameters. Some examples connect inputs of the translated NN to upstream operators of the ML pipeline that had not been translated. Some examples further tune the translated NN using backpropagation. Some examples determine whether an operator is trainable or non-trainable and flag related parameters accordingly for later training. Some examples determine whether an operator has multiple corresponding NN modules within the translation dictionary and make an optimized selection.

公开(公告)号：US20210124739A1

公开(公告)日：2021-04-29

申请号：US16990506

申请日：2020-08-11

申请人： Microsoft Technology Licensing, LLC

发明人： Konstantinos KARANASOS ,  Matteo INTERLANDI ,  Fotios PSALLIDAS ,  Rathijit SEN ,  Kwanghyun PARK ,  Ivan POPIVANOV ,  Subramaniam VENKATRAMAN KRISHNAN ,  Markus WEIMER ,  Yuan YU ,  Raghunath RAMAKRISHNAN ,  Carlo Aldo CURINO ,  Doris Suiyi XIN ,  Karla Jean SAUR

IPC分类号： G06F16/2458 ,  G06N5/04 ,  G06N20/00 ,  G06F16/28

摘要： The description relates to executing an inference query relative to a database management system, such as a relational database management system. In one example a trained machine learning model can be stored within the database management system. An inference query can be received that applies the trained machine learning model on data local to the database management system. Analysis can be performed on the inference query and the trained machine learning model to generate a unified intermediate representation of the inference query and the trained model. Cross optimization can be performed on the unified intermediate representation. Based upon the cross-optimization, a first portion of the unified intermediate representation to be executed by a database engine of the database management system can be determined, and, a second portion of the unified intermediate representation to be executed by a machine learning runtime can be determined.

公开(公告)号：US20230342359A1

公开(公告)日：2023-10-26

申请号：US18345789

申请日：2023-06-30

申请人： Microsoft Technology Licensing, LLC

发明人： Irene Rogan SHAFFER ,  Remmelt Herbert Lieve AMMERLAAN ,  Gilbert ANTONIUS ,  Marc T. FRIEDMAN ,  Abhishek ROY ,  Lucas ROSENBLATT ,  Vijay Kumar RAMANI ,  Shi QIAO ,  Alekh JINDAL ,  Peter ORENBERG ,  H M Sajjad Hossain ,  Soundararajan Srinivasan ,  Hiren Shantilal PATEL ,  Markus WEIMER

IPC分类号： G06F16/2453 ,  G06N20/00 ,  G06F11/34 ,  G06F16/901

CPC分类号： G06F16/24542 ,  G06N20/00 ,  G06F11/3466 ,  G06F16/9024

摘要： Methods of machine learning for system deployments without performance regressions are performed by systems and devices. A performance safeguard system is used to design pre-production experiments for determining the production readiness of learned models based on a pre-production budget by leveraging big data processing infrastructure and deploying a large set of learned or optimized models for its query optimizer. A pipeline for learning and training differentiates the impact of query plans with and without the learned or optimized models, selects plan differences that are likely to lead to most dramatic performance difference, runs a constrained set of pre-production experiments to empirically observe the runtime performance, and finally picks the models that are expected to lead to consistently improved performance for deployment. The performance safeguard system enables safe deployment not just for learned or optimized models but also for additional of other ML-for-Systems features.

公开(公告)号：US20240232634A1

公开(公告)日：2024-07-11

申请号：US18423254

申请日：2024-01-25

申请人： Microsoft Technology Licensing, LLC

发明人： Matteo INTERLANDI ,  Byung-Gon CHUN ,  Markus WEIMER ,  Gyeongin YU ,  Saeed AMIZADEH

IPC分类号： G06N3/084 ,  G06F16/901 ,  G06N20/00 ,  G06T1/20

CPC分类号： G06N3/084 ,  G06F16/9027 ,  G06N20/00 ,  G06T1/20

摘要： Solutions for adapting machine learning (ML) models to neural networks (NNs) include receiving an ML pipeline comprising a plurality of operators; determining operator dependencies within the ML pipeline; determining recognized operators; for each of at least two recognized operators, selecting a corresponding NN module from a translation dictionary; and wiring the selected NN modules in accordance with the operator dependencies to generate a translated NN. Some examples determine a starting operator for translation, which is the earliest recognized operator having parameters. Some examples connect inputs of the translated NN to upstream operators of the ML pipeline that had not been translated. Some examples further tune the translated NN using backpropagation. Some examples determine whether an operator is trainable or non-trainable and flag related parameters accordingly for later training. Some examples determine whether an operator has multiple corresponding NN modules within the translation dictionary and make an optimized selection.