Search

Xiaodong He

from Sammamish, WA
Age ~51
Get Report
Also known as:
Xiaodong He Ma, Xiao Dong He
Phone and address:
1580 237Th Ct NE, Redmond, WA 98074
Related to:
Moping LiJiafan Liu, 62
Connected to:
Aaron Y He, 37Airel He, 63Bertin P He, 33Bill He, 61Chenchen He, 42Christa Y He, 82Chuanjie He, 34Da G He, 65Daofeng He, 68Dawei He, 37

Xiaodong He Phones & Addresses

  • 1580 237Th Ct NE, Sammamish, WA 98074
  • Redmond, WA
  • Columbia, MO
  • Issaquah, WA

Resumes

Resumes

Xiaodong He Photo 1

Deputy Managing Director Of Jd Ai Research, Technical Vice Preseident

View page
Location:
Seattle, WA
Industry:
Computer Software
Work:
Jd.com
Deputy Managing Director of Jd Ai Research, Technical Vice Preseident
Microsoft 2017 - 2018
Principal Researcher and Research Manager
Microsoft 2006 - 2018
Principal Researcher and Affiliate Professor at U of Washington
University of Washington 2006 - 2018
Affiliate Professor
Microsoft Sep 2003 - May 2006
Software Development Engineer
Education:
University of Missouri - Columbia 1999 - 2003
Doctorates, Doctor of Philosophy, Computer Science Chinese Academy of Sciences 1996 - 1999
Masters Tsinghua University 1991 - 1996
Bachelors Chinese Academy of Sciences 1982 - 1984
Masters
Skills:
Natural Language Processing
Machine Learning
Pattern Recognition
Artificial Intelligence
Speech Recognition
Information Retrieval
Data Mining
Text Mining
Information Extraction
Software Development
Computational Linguistics
Machine Translation
Java
Computer Vision
Signal Processing
Big Data
Mapreduce
Human Computer Interaction
Languages:
English
Mandarin
Xiaodong He Photo 2

Researcher At Microsoft Research - Redmond

View page
Position:
Researcher at Microsoft, Affiliate Professor at University of Washington
Location:
Greater Seattle Area
Industry:
Computer Software
Work:
Microsoft
Researcher
University of Washington - Greater Seattle Area since 2012
Affiliate Professor
Avaya Jun 2001 - Sep 2001
Intern
Education:
University of Missouri-Columbia 1999 - 2003
PhD, Computer Science Chinese Academy of Sciences 1996 - 1999
Master, Singal and Information Processing Tsinghua University 1991 - 1996
Bachelor, Precision Instruments
Skills:
Natural Language Processing
Machine Learning
Speech Recognition
Information Retrieval
Pattern Recognition
Text Mining
Computational Linguistics
Information Extraction
Artificial Intelligence
Data Mining
Machine Translation
Languages:
English
Chinese

Publications

Us Patents

Speech Models Generated Using Competitive Training, Asymmetric Training, And Data Boosting

View page
US Patent:
7693713, Apr 6, 2010
Filed:
Jun 17, 2005
Appl. No.:
11/156106
Inventors:
Xiaodong He - Issaquah WA, US
Jian Wu - Redmond WA, US
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G10L 15/06
US Classification:
704243, 704245
Abstract:
Speech models are trained using one or more of three different training systems. They include competitive training which reduces a distance between a recognized result and a true result, data boosting which divides and weights training data, and asymmetric training which trains different model components differently.

Segment-Discriminating Minimum Classification Error Pattern Recognition

View page
US Patent:
7873209, Jan 18, 2011
Filed:
Jan 31, 2007
Appl. No.:
11/700664
Inventors:
Li Deng - Sammamish WA, US
Xiaodong He - Issaquah WA, US
Qiang Fu - Atlanta GA, US
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06K 9/00
G10L 15/00
US Classification:
382159, 382181, 704231
Abstract:
Pattern model parameters are updated using update equations based on competing patterns that are identical to a reference pattern except for one segment at a time that is replaced with a competing segment. This allows pattern recognition parameters to be tuned one segment at a time, rather than have to try to model distinguishing features of the correct pattern model as a whole, according to an illustrative embodiment. A reference pattern and competing patterns are divided into pattern segments. A set of training patterns are generated by replacing one of the pattern segments in the reference pattern with a corresponding competing pattern segment. For each of the training patterns, a pattern recognition model is applied to evaluate a relative degree of correspondence of the reference pattern with the pattern signal compared to a degree of correspondence of the training patterns with the pattern signal.

Hmm Alignment For Combining Translation Systems

View page
US Patent:
8060358, Nov 15, 2011
Filed:
Jun 27, 2008
Appl. No.:
12/147807
Inventors:
Xiaodong He - Issaquah WA, US
Mei Yang - Seattle WA, US
Jianfeng Gao - Kirkland WA, US
Patrick Nguyen - Seattle WA, US
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06F 17/28
US Classification:
704 2, 704277, 704 9, 704256
Abstract:
A computing system configured to produce an optimized translation hypothesis of text input into the computing system. The computing system includes a plurality of translation machines. Each of the translation machines is configured to produce their own translation hypothesis from the same text. An optimization machine is connected to the plurality of translation machines. The optimization machine is configured to receive the translation hypotheses from the translation machines. The optimization machine is further configured to align, word-to-word, the hypotheses in the plurality of hypotheses by using a hidden Markov model.

Word-Dependent Transition Models In Hmm Based Word Alignment For Statistical Machine Translation

View page
US Patent:
8060360, Nov 15, 2011
Filed:
Oct 30, 2007
Appl. No.:
11/980257
Inventors:
Xiaodong He - Issaquah WA, US
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06F 17/27
US Classification:
704 9, 704 10, 704 2
Abstract:
A word alignment modeler uses probabilistic learning techniques to train “word-dependent transition models” for use in constructing phrase level Hidden Markov Model (HMM) based word alignment models. As defined herein, “word-dependent transition models” provide a probabilistic model wherein for each source word in training data, a self-transition probability is modeled in combination with a probability of jumping from that particular word to a different word, thereby providing a full transition model for each word in a source phrase. HMM based word alignment models are then used for various word alignment and machine translation tasks. In additional embodiments sparse data problems (i. e. , rarely used words) are addressed by using probabilistic learning techniques to estimate word-dependent transition model parameters by maximum a posteriori (MAP) training.

Identifying Language Origin Of Words

View page
US Patent:
8185376, May 22, 2012
Filed:
Mar 20, 2006
Appl. No.:
11/384401
Inventors:
Min Chu - Beijing, CN
Yi Ning Chen - Beijing, CN
Xiaodong He - Issaquah WA, US
Megan Riley - Kirkland WA, US
Kevin E. Feige - Duvall WA, US
Yifan Gong - Sammamish WA, US
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06F 17/20
G06F 17/27
US Classification:
704 8, 704 9
Abstract:
The language of origin of a word is determined by analyzing non-uniform letter sequence portions of the word.

Minimum Classification Error Training With Growth Transformation Optimization

View page
US Patent:
8301449, Oct 30, 2012
Filed:
Oct 16, 2006
Appl. No.:
11/581673
Inventors:
Xiaodong He - Issaquah WA, US
Li Deng - Redmond WA, US
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G10L 15/00
US Classification:
704257, 704251, 704256, 7042561, 7042562, 7042564, 7042566
Abstract:
Hidden Markov Model (HMM) parameters are updated using update equations based on growth transformation optimization of a minimum classification error objective function. Using the list of N-best competitor word sequences obtained by decoding the training data with the current-iteration HMM parameters, the current HMM parameters are updated iteratively. The updating procedure involves using weights for each competitor word sequence that can take any positive real value. The updating procedure is further extended to the case where a decoded lattice of competitors is used. In this case, updating the model parameters relies on determining the probability for a state at a time point based on the word that spans the time point instead of the entire word sequence. This word-bound span of time is shorter than the duration of the entire word sequence and thus reduces the computing time.

Confidence Threshold Tuning

View page
US Patent:
8396715, Mar 12, 2013
Filed:
Jun 28, 2005
Appl. No.:
11/168278
Inventors:
Julian J. Odell - Kirkland WA, US
Li Jiang - Redmond WA, US
Wei Zhang - Kirkland WA, US
Xiaodong He - Issaquah WA, US
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G10L 21/00
G10L 15/00
US Classification:
704270, 704231, 7042701
Abstract:
An expected dialog-turn (ED) value is estimated for evaluating a speech application. Parameters such as a confidence threshold setting can be adjusted based on the expected dialog-turn value. In a particular example, recognition results and corresponding confidence scores are used to estimate the expected dialog-turn value. The recognition results can be associated with a possible outcome for the speech application and a cost for the possible outcome can be used to estimate the expected dialog-turn value.

Integrative And Discriminative Technique For Spoken Utterance Translation

View page
US Patent:
8407041, Mar 26, 2013
Filed:
Dec 1, 2010
Appl. No.:
12/957394
Inventors:
Li Deng - Redmond WA, US
Yaodong Zhang - Boston MA, US
Alejandro Acero - Bellevue WA, US
Xiaodong He - Issaquah WA, US
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06F 17/28
US Classification:
704 2
Abstract:
Architecture that provides the integration of automatic speech recognition (ASR) and machine translation (MT) components of a full speech translation system. The architecture is an integrative and discriminative approach that employs an end-to-end objective function (the conditional probability of the translated sentence (target) given the source language's acoustic signal, as well as the associated BLEU score in the translation, as a goal in the integrated system. This goal defines the theoretically correct variables to determine the speech translation system output using a Bayesian decision rule. These theoretically correct variables are modified in practical use due to known imperfections of the various models used in building the full speech translation system. The disclosed approach also employs automatic training of these variables using minimum classification error (MCE) criterion. The measurable BLEU scores are used to facilitate the implementation of the MCE training procedure in a step that defines the class-specific discriminant function.
Control profile
Xiaodong He from Sammamish, WA, age ~51 Get Report