Microsoft Technology Licensing, LLC

Other

Chinese Academy of Sciences

Jianfeng Gao

Xiaodong He

Li Deng

Xinying Song

Jinchao Li

Pengcheng He

Ming Zhou

Yu Wang

William B Dolan

Xiaodong Liu

Jianshu Chen

Prabhdeep Singh

Xiujun Li

Hui Su

Yelong Shen

Mingjing Li

Ye-Yi Wang

Christopher John Brockett

Weizhu Chen

Chang-Ning Huang

Haiyuan Cao

Zheng Chen

Dilek Zeynep Hakkani-Tur

Gokhan Tur

Kai-Fu Lee

Jaideep Sarkar

Lihong Li

Margaret Ann Mitchell

William Ramsey

Christopher Brian Quirk

Chenlei Guo

Sanjeev Katariya

Mu Li

Michel Galley

Alessandro Sordoni

Asli Celikyilmaz

Michael Auli

Jian-Yun Nie

Alejandro Acero

Xuedong D Huang

Hang Li

Hua Wu

Kuansan Wang

Larry Paul Heck

Yue Hua Zhang

Michael Gamon

Hsiao-Wuen Hon

Nanshan Zeng

Ruofei Zhang

Yunbo Cao

Sanaz Ahari

Lars Hasso Liden

Hisami Suzuki

Kristina Nikolova Toutanova

Hoifung Poon

Wen-tau Yih

Gu Xu

Brian D Remick

Qi Yao

Chenguang Zhu

Thomas Park

Alnur Ali

Zhenghao Wang

Lei Zhang

Qiang Wu

Jian Qiao Sun

Feng Zhang

John Carlton Platt

Joshua T Goodman

Chao Zhao

Jun Liu

Ji-Rong Wen

Krysta Marie Svore

Paul Nathan Bennett

Rui Yang

Gregory Nicholas Hullender

Junfeng Li

Geoffrey G Zweig

Mei-Yuh Hwang

Patrick Pantel

Guilei Wang

Gary Kacmarcik

LuMing Wang

Paul Anthony Crook

Chin-Yew Lin

Shahin Shayandeh

Zijian Zheng

Vikram Reddy Dendi

Yuxiong He

Pallavi Choudhury

Jinbiao Liu

Jian Jiao

Guihong Cao

Janine Crumb

Guowei Liu

Patrick Nguyen

Elbio Renato Torres Abib

Achraf Chalabi

Chris J C Burges

Systems and method are provided that are directed to tuning a hyperparameter associated with a small neural network model and transferring the hyperparameter to a large neural network model. At least one neural network model may be received along with a request for one or more tuned hyperparameters. Prior to scaling the large neural network, the large neural network is parameterized in accordance with a parameterizing scheme. The large neural network is then scaled and reduced in size such that a hyperparameter tuning process may be performed. A tuned hyperparameter may then be provided to a requestor such that the hyperparameter can be directly input into the large neural network. By tuning a hyper parameter using a small neural network, significant computation cycles and energy may be saved.

Hyperparameter transfer via the theory of infinite-width neural networks

This document relates to training of machine learning models. One example method involves providing a machine learning model having one or more mapping layers. The one or more mapping layers can include at least a first mapping layer configured to map components of pretraining examples into first representations in a space. The example method also includes performing a pretraining stage on the one or more mapping layers using the pretraining examples. The pretraining stage can include adding noise to the first representations of the components of the pretraining examples to obtain noise-adjusted first representations. The pretraining stage can also include performing a self-supervised learning process to pretrain the one or more mapping layers using at least the first representations of the training data items and the noise-adjusted first representations of the training data items.

Adversarial pretraining of machine learning models

This document relates to training of machine learning models such as neural networks. One example method involves providing a machine learning model having one or more layers and associated parameters and performing a pretraining stage on the parameters of the machine learning model to obtain pretrained parameters. The example method also involves performing a tuning stage on the machine learning model by using labeled training samples to tune the pretrained parameters. The tuning stage can include performing noise adjustment of the labeled training examples to obtain noise-adjusted training samples. The tuning stage can also include adjusting the pretrained parameters based at least on the labeled training examples and the noise-adjusted training examples to obtain adapted parameters. The example method can also include outputting a tuned machine learning model having the adapted parameters.

Adversarial training of machine learning models

A method for training a language model comprises (a) receiving vectorized training data as input to a multitask pretraining problem; (b) generating modified vectorized training data based on the vectorized training data, according to an upstream data embedding; (c) emitting pretraining output based on the modified vectorized training data, according to a downstream data embedding equivalent to the upstream data embedding; and (d) adjusting the upstream data embedding and the downstream data embedding by computing, based on the pretraining output, a gradient of the upstream data embedding disentangled from a gradient of the downstream data embedding, thereby advancing the multitask pretraining problem toward a pretrained state.

Language-model pretraining with gradient-disentangled embedding sharing

Systems and methods are provided for facilitating the building and use of natural language understanding models. The systems and methods identify a plurality of tokens and use them to generate one or more pre-trained natural language models using a transformer. The transformer disentangles the content embedding and positional embedding in the computation of its attention matrix. Systems and methods are also provided to facilitate self-training of the pre-trained natural language model by utilizing multi-step decoding to better reconstruct masked tokens and improve pre-training convergence.

Efficient transformer language models with disentangled attention and multi-step decoding

Systems and methods are provided for determining a response to a query in a dialog. An entity extractor extracts rules and conditions associated with the query and determines a particular task. The disclosed technology generates a transformer-based dialog embedding by pre-training a transformer using dialog corpora including a plurality of tasks. A task-specific classifier generates a first set of candidate responses based on rules and conditions associated with the task. The transformer-based dialog embedding generates a second set of candidate responses to the query. The classifier accommodates changes made to a task by an interactive dialog editor as machine teaching. A response generator generates a response based on the first and second sets of candidate responses using an optimization function. The disclosed technology leverages both a data-driven, generative model (a transformer) based on dialog corpora and a user-driven, task-specific rule-based classifier that accommodating updates in rules and conditions associated with a particular task.

Hybrid transformer-based dialog processor

This document relates to architectures and training procedures for multi-task machine learning models, such as neural networks. One example method involves providing a multi-task machine learning model having one or more shared layers and two or more task-specific layers. The method can also involve performing a pretraining stage on the one or more shared layers using one or more unsupervised prediction tasks. The method can also involve performing a tuning stage on the one or more shared layers and the two or more task-specific layers using respective task-specific objectives.

Multi-task machine learning architectures and training procedures

Methods and systems are disclosed for improving dialog management for task-oriented dialog systems. The disclosed dialog builder leverages machine teaching processing to improve development of dialog managers. In this way, the dialog builder combines the strengths of both rule-based and machine-learned approaches to allow dialog authors to: (1) import a dialog graph developed using popular dialog composers, (2) convert the dialog graph to text-based training dialogs, (3) continuously improve the trained dialogs based on log dialogs, and (4) generate a corrected dialog for retraining the machine learning.

Training a user-system dialog in a task-oriented dialog system

This document relates to machine learning. One example includes a method or technique that can be performed on a computing device. The method or technique can include obtaining a task-semantically-conditioned generative model that has been pretrained based at least on a first training data set having unlabeled training examples and semantically conditioned based at least on a second training data set having dialog act-labeled utterances. The method or technique can also include inputting dialog acts into the semantically-conditioned generative model and obtaining synthetic utterances that are output by the semantically-conditioned generative model. The method or technique can also include outputting the synthetic utterances.

Synthetic data generation for training of natural language understanding models

Generally discussed herein are devices, systems, and methods for generating an embedding that is both local string dependent and global string dependent. The generated embedding can improve machine learning (ML) model performance. A method can include converting a string of words to a series of tokens, generating a local string-dependent embedding of each token of the series of tokens, generating a global string-dependent embedding of each token of the series of tokens, combining the local string dependent embedding the global string dependent embedding to generate an n-gram induced embedding of each token of the series of tokens, obtaining a masked language model (MLM) previously trained to generate a masked word prediction, and executing the MLM based on the n-based induced embedding of each token to generate the masked word prediction.

ML using n-gram induced input representation

A processing unit can train a model as a joint multi-domain recurrent neural network (JRNN), such as a bi-directional recurrent neural network (bRNN) and/or a recurrent neural network with long-short term memory (RNN-LSTM) for spoken language understanding (SLU). The processing unit can use the trained model to, e.g., jointly model slot filling, intent determination, and domain classification. The joint multi-domain model described herein can estimate a complete semantic frame per query, and the joint multi-domain model enables multi-task deep learning leveraging the data from multiple domains. The joint multi-domain recurrent neural network (JRNN) can leverage semantic intents (such as, finding or identifying, e.g., a domain specific goal) and slots (such as, dates, times, locations, subjects, etc.) across multiple domains.

Multi-domain joint semantic frame parsing

Generally discussed herein are devices, systems, and methods for scheduling tasks to be completed by resources. A method can include identifying features of the task, the features including a time-dependent feature and a time-independent feature, the time-dependent feature indicating a time the task is more likely to be successfully completed by the resource, converting the features to feature values based on a predefined mapping of features to feature values in a first memory device, determining, by a gradient boost tree model and based on a first current time and the feature values, a likelihood the resource will successfully complete the task, and scheduling the task to be performed by the resource based on the determined likelihood.

Resource scheduling using machine learning

Techniques for iterative query-based analysis of text are described. According to various implementations, a neural network architecture is implemented receives a query for information about text content, and iteratively analyzes the content using the query. During the analysis a state of the query evolves until it reaches a termination state, at which point the state of the query is output as an answer to the initial query.

Iterative query-based analysis of text

Knowledge graphs can greatly improve the quality of content recommendation systems. There is a broad variety of knowledge graphs in the domain including clicked user-ad graphs, clicked query-ad graphs, keyword-display URL graphs etc. A hierarchical Transformer model learns entity embeddings in knowledge graphs. The model consists of two different Transformer blocks where the bottom block generates relation-dependent embeddings for the source entity and its neighbors, and the top block aggregates the outputs from the bottom block to produce the target entity embedding. To balance the information from contextual entities and the source entity itself, a masked entity model (MEM) task is combined with a link prediction task in model training.

Learning graph representations using hierarchical transformers for content recommendation

Growing community of inventors

Woodinville, WA, United States of America

Jianfeng Gao