SlideShare a Scribd company logo
Time:11:00AM

Location:New CSIE Building R110

Topic:Learning with Integer Linear Programming Inference for Constrained Output

Speaker: Scott Wen-Tau Yih

Abstract:

       In several structured classification problems, explicit and expressive constraints are crucial to

enhancing the accuracy and quality of the predictions. However, it was not clear how this additional

knowledge can be used in various learning frameworks. In this talk, I'll first demonstrate how constraints

can be incorporated in Conditional Random Fields via a novel inference approach based on integer

linear programming. Inference in CRFs and HMMs is usually done using the Viterbi algorithm, an

efficient dynamic programming algorithm. In many cases, general (non-local and non-sequential)

constraints may exist over the output sequence, but cannot be incorporated and exploited in a natural

way by Viterbi. Our inference procedure extends CRF models to naturally and efficiently support general

constraint structures. For sequential constraints, this procedure reduces to simple linear programming
as the inference process. Experimental evidences of our approach will be provided in the context of an

important NLP problem, semantic role labeling.

       One interesting phenomenon we observed in the experiments is that a simple learning plus

inference scheme may outperform inference based training approaches when incorporating constraints.

In the second part of my talk, I'll describe how we compared these two learning frameworks by

observing their behaviors in different conditions. Experiments and theoretical justification lead to the

conclusion that using inference based learning is superior when the local classifiers are difficult to learn

but may require many examples before any discernible difference can be observed.

Bio:

       Wen-tau Yih is a post-doc researcher in the Machine Learning and Applied Statistics group at

Microsoft Research. He got his Ph.D. at the University of Illinois at Urbana-Champaign in May 2005.

Although his current research focuses mainly on problems related to email applications and anti-spam,

his research interests spread on various problems in Machine Learning and Natural Language

Processing, such as learning and knowledge representation, information extraction, semantic parsing,

and inference and learning for structured output. Wen-tau received both his M.S. and B.S. degrees in

Computer Science from National Taiwan University. More information can be found on his homepage:

http://scottyih.org/

More Related Content

What's hot

Software Engineering Ontology and Software Testing
Software Engineering Ontology and Software Testing�Software Engineering Ontology and Software Testing�
Software Engineering Ontology and Software Testing
Kamal Patel
 
An Overview of Noise-Robust Automatic Speech Recognition
An Overview of Noise-Robust Automatic Speech RecognitionAn Overview of Noise-Robust Automatic Speech Recognition
An Overview of Noise-Robust Automatic Speech Recognition
Projectsatbangalore
 
Audit report[rollno 49]
Audit report[rollno 49]Audit report[rollno 49]
Audit report[rollno 49]
RAHULROHAM2
 
Association Rule Mining Based Extraction of Semantic Relations Using Markov L...
Association Rule Mining Based Extraction of Semantic Relations Using Markov L...Association Rule Mining Based Extraction of Semantic Relations Using Markov L...
Association Rule Mining Based Extraction of Semantic Relations Using Markov L...
IJwest
 
Supervised Approach to Extract Sentiments from Unstructured Text
Supervised Approach to Extract Sentiments from Unstructured TextSupervised Approach to Extract Sentiments from Unstructured Text
Supervised Approach to Extract Sentiments from Unstructured Text
International Journal of Engineering Inventions www.ijeijournal.com
 
Improving Robustness and Flexibility of Concept Taxonomy Learning from Text
Improving Robustness and Flexibility of Concept Taxonomy Learning from Text Improving Robustness and Flexibility of Concept Taxonomy Learning from Text
Improving Robustness and Flexibility of Concept Taxonomy Learning from Text
University of Bari (Italy)
 
Machine Learning
Machine LearningMachine Learning
Machine Learning
butest
 
Ontological Model of Educational Programs in Computer Science (Bachelor and M...
Ontological Model of Educational Programs in Computer Science (Bachelor and M...Ontological Model of Educational Programs in Computer Science (Bachelor and M...
Ontological Model of Educational Programs in Computer Science (Bachelor and M...
ijsrd.com
 
Classification and Regression
Classification and RegressionClassification and Regression
Classification and Regression
Megha Sharma
 
810 research proposal
810 research proposal810 research proposal
810 research proposal
kpatric1
 
M phil
M philM phil
M phil
peeroz
 
GRAPHICAL REPRESENTATION IN TUTORING SYSTEMS
GRAPHICAL REPRESENTATION IN TUTORING SYSTEMSGRAPHICAL REPRESENTATION IN TUTORING SYSTEMS
GRAPHICAL REPRESENTATION IN TUTORING SYSTEMS
ijcsit
 
Design pattern 1
Design pattern 1Design pattern 1
Design pattern 1
Naga Muruga
 
Machine learning
Machine learningMachine learning
Machine learning
hplap
 
University-Toronto-Program-The-FundamentalsV2
University-Toronto-Program-The-FundamentalsV2University-Toronto-Program-The-FundamentalsV2
University-Toronto-Program-The-FundamentalsV2
Majid Hameed
 

What's hot (15)

Software Engineering Ontology and Software Testing
Software Engineering Ontology and Software Testing�Software Engineering Ontology and Software Testing�
Software Engineering Ontology and Software Testing
 
An Overview of Noise-Robust Automatic Speech Recognition
An Overview of Noise-Robust Automatic Speech RecognitionAn Overview of Noise-Robust Automatic Speech Recognition
An Overview of Noise-Robust Automatic Speech Recognition
 
Audit report[rollno 49]
Audit report[rollno 49]Audit report[rollno 49]
Audit report[rollno 49]
 
Association Rule Mining Based Extraction of Semantic Relations Using Markov L...
Association Rule Mining Based Extraction of Semantic Relations Using Markov L...Association Rule Mining Based Extraction of Semantic Relations Using Markov L...
Association Rule Mining Based Extraction of Semantic Relations Using Markov L...
 
Supervised Approach to Extract Sentiments from Unstructured Text
Supervised Approach to Extract Sentiments from Unstructured TextSupervised Approach to Extract Sentiments from Unstructured Text
Supervised Approach to Extract Sentiments from Unstructured Text
 
Improving Robustness and Flexibility of Concept Taxonomy Learning from Text
Improving Robustness and Flexibility of Concept Taxonomy Learning from Text Improving Robustness and Flexibility of Concept Taxonomy Learning from Text
Improving Robustness and Flexibility of Concept Taxonomy Learning from Text
 
Machine Learning
Machine LearningMachine Learning
Machine Learning
 
Ontological Model of Educational Programs in Computer Science (Bachelor and M...
Ontological Model of Educational Programs in Computer Science (Bachelor and M...Ontological Model of Educational Programs in Computer Science (Bachelor and M...
Ontological Model of Educational Programs in Computer Science (Bachelor and M...
 
Classification and Regression
Classification and RegressionClassification and Regression
Classification and Regression
 
810 research proposal
810 research proposal810 research proposal
810 research proposal
 
M phil
M philM phil
M phil
 
GRAPHICAL REPRESENTATION IN TUTORING SYSTEMS
GRAPHICAL REPRESENTATION IN TUTORING SYSTEMSGRAPHICAL REPRESENTATION IN TUTORING SYSTEMS
GRAPHICAL REPRESENTATION IN TUTORING SYSTEMS
 
Design pattern 1
Design pattern 1Design pattern 1
Design pattern 1
 
Machine learning
Machine learningMachine learning
Machine learning
 
University-Toronto-Program-The-FundamentalsV2
University-Toronto-Program-The-FundamentalsV2University-Toronto-Program-The-FundamentalsV2
University-Toronto-Program-The-FundamentalsV2
 

Similar to 20051128.doc

Nlp research presentation
Nlp research presentationNlp research presentation
Nlp research presentation
Surya Sg
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
Sri Ambati
 
Survey on contrastive self supervised l earning
Survey on contrastive self supervised l earningSurvey on contrastive self supervised l earning
Survey on contrastive self supervised l earning
Anirudh Ganguly
 
DALL-E 2 - OpenAI imagery automation first developed by Vishal Coodye in 2021...
DALL-E 2 - OpenAI imagery automation first developed by Vishal Coodye in 2021...DALL-E 2 - OpenAI imagery automation first developed by Vishal Coodye in 2021...
DALL-E 2 - OpenAI imagery automation first developed by Vishal Coodye in 2021...
MITAILibrary
 
IMBALANCED DATASET EFFECT ON CNN-BASED CLASSIFIER PERFORMANCE FOR FACE RECOGN...
IMBALANCED DATASET EFFECT ON CNN-BASED CLASSIFIER PERFORMANCE FOR FACE RECOGN...IMBALANCED DATASET EFFECT ON CNN-BASED CLASSIFIER PERFORMANCE FOR FACE RECOGN...
IMBALANCED DATASET EFFECT ON CNN-BASED CLASSIFIER PERFORMANCE FOR FACE RECOGN...
gerogepatton
 
Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recogn...
Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recogn...Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recogn...
Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recogn...
gerogepatton
 
HyperQA: A Framework for Complex Question-Answering
HyperQA: A Framework for Complex Question-AnsweringHyperQA: A Framework for Complex Question-Answering
HyperQA: A Framework for Complex Question-Answering
Jinho Choi
 
Association Rule Mining Based Extraction of Semantic Relations Using Markov ...
Association Rule Mining Based Extraction of  Semantic Relations Using Markov ...Association Rule Mining Based Extraction of  Semantic Relations Using Markov ...
Association Rule Mining Based Extraction of Semantic Relations Using Markov ...
dannyijwest
 
CMPE258 Short story.pptx
CMPE258 Short story.pptxCMPE258 Short story.pptx
CMPE258 Short story.pptx
ChirudeepGorle
 
Text classification supervised algorithms with term frequency inverse documen...
Text classification supervised algorithms with term frequency inverse documen...Text classification supervised algorithms with term frequency inverse documen...
Text classification supervised algorithms with term frequency inverse documen...
IJECEIAES
 
A Formal Machine Learning or Multi Objective Decision Making System for Deter...
A Formal Machine Learning or Multi Objective Decision Making System for Deter...A Formal Machine Learning or Multi Objective Decision Making System for Deter...
A Formal Machine Learning or Multi Objective Decision Making System for Deter...
Editor IJCATR
 
Lagging_Inference_Networks_and_Posterior_Collapse_.pdf
Lagging_Inference_Networks_and_Posterior_Collapse_.pdfLagging_Inference_Networks_and_Posterior_Collapse_.pdf
Lagging_Inference_Networks_and_Posterior_Collapse_.pdf
AnkitBiswas31
 
Tracing Requirements as a Problem of Machine Learning
Tracing Requirements as a Problem of Machine Learning Tracing Requirements as a Problem of Machine Learning
Tracing Requirements as a Problem of Machine Learning
ijseajournal
 
Meetup 22/2/2018 - Artificiële Intelligentie & Human Resources
Meetup 22/2/2018 - Artificiële Intelligentie & Human ResourcesMeetup 22/2/2018 - Artificiële Intelligentie & Human Resources
Meetup 22/2/2018 - Artificiële Intelligentie & Human Resources
Digipolis Antwerpen
 
Re2018 Semios for Requirements
Re2018 Semios for RequirementsRe2018 Semios for Requirements
Re2018 Semios for Requirements
Clément Portet
 
NE7012- SOCIAL NETWORK ANALYSIS
NE7012- SOCIAL NETWORK ANALYSISNE7012- SOCIAL NETWORK ANALYSIS
NE7012- SOCIAL NETWORK ANALYSIS
rathnaarul
 
Report of Previous Project by Yifan Guo
Report of Previous Project by Yifan GuoReport of Previous Project by Yifan Guo
Report of Previous Project by Yifan Guo
Yifan Guo
 
A Review of Constraint Programming
A Review of Constraint ProgrammingA Review of Constraint Programming
A Review of Constraint Programming
Editor IJCATR
 
Access To Specific Declarative Knowledge By Expert Systems The Impact Of Log...
Access To Specific Declarative Knowledge By Expert Systems  The Impact Of Log...Access To Specific Declarative Knowledge By Expert Systems  The Impact Of Log...
Access To Specific Declarative Knowledge By Expert Systems The Impact Of Log...
Audrey Britton
 
Are Human-generated Demonstrations Necessary for In-context Learning?
Are Human-generated Demonstrations Necessary for In-context Learning?Are Human-generated Demonstrations Necessary for In-context Learning?
Are Human-generated Demonstrations Necessary for In-context Learning?
MENGSAYLOEM1
 

Similar to 20051128.doc (20)

Nlp research presentation
Nlp research presentationNlp research presentation
Nlp research presentation
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
 
Survey on contrastive self supervised l earning
Survey on contrastive self supervised l earningSurvey on contrastive self supervised l earning
Survey on contrastive self supervised l earning
 
DALL-E 2 - OpenAI imagery automation first developed by Vishal Coodye in 2021...
DALL-E 2 - OpenAI imagery automation first developed by Vishal Coodye in 2021...DALL-E 2 - OpenAI imagery automation first developed by Vishal Coodye in 2021...
DALL-E 2 - OpenAI imagery automation first developed by Vishal Coodye in 2021...
 
IMBALANCED DATASET EFFECT ON CNN-BASED CLASSIFIER PERFORMANCE FOR FACE RECOGN...
IMBALANCED DATASET EFFECT ON CNN-BASED CLASSIFIER PERFORMANCE FOR FACE RECOGN...IMBALANCED DATASET EFFECT ON CNN-BASED CLASSIFIER PERFORMANCE FOR FACE RECOGN...
IMBALANCED DATASET EFFECT ON CNN-BASED CLASSIFIER PERFORMANCE FOR FACE RECOGN...
 
Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recogn...
Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recogn...Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recogn...
Imbalanced Dataset Effect on CNN-Based Classifier Performance for Face Recogn...
 
HyperQA: A Framework for Complex Question-Answering
HyperQA: A Framework for Complex Question-AnsweringHyperQA: A Framework for Complex Question-Answering
HyperQA: A Framework for Complex Question-Answering
 
Association Rule Mining Based Extraction of Semantic Relations Using Markov ...
Association Rule Mining Based Extraction of  Semantic Relations Using Markov ...Association Rule Mining Based Extraction of  Semantic Relations Using Markov ...
Association Rule Mining Based Extraction of Semantic Relations Using Markov ...
 
CMPE258 Short story.pptx
CMPE258 Short story.pptxCMPE258 Short story.pptx
CMPE258 Short story.pptx
 
Text classification supervised algorithms with term frequency inverse documen...
Text classification supervised algorithms with term frequency inverse documen...Text classification supervised algorithms with term frequency inverse documen...
Text classification supervised algorithms with term frequency inverse documen...
 
A Formal Machine Learning or Multi Objective Decision Making System for Deter...
A Formal Machine Learning or Multi Objective Decision Making System for Deter...A Formal Machine Learning or Multi Objective Decision Making System for Deter...
A Formal Machine Learning or Multi Objective Decision Making System for Deter...
 
Lagging_Inference_Networks_and_Posterior_Collapse_.pdf
Lagging_Inference_Networks_and_Posterior_Collapse_.pdfLagging_Inference_Networks_and_Posterior_Collapse_.pdf
Lagging_Inference_Networks_and_Posterior_Collapse_.pdf
 
Tracing Requirements as a Problem of Machine Learning
Tracing Requirements as a Problem of Machine Learning Tracing Requirements as a Problem of Machine Learning
Tracing Requirements as a Problem of Machine Learning
 
Meetup 22/2/2018 - Artificiële Intelligentie & Human Resources
Meetup 22/2/2018 - Artificiële Intelligentie & Human ResourcesMeetup 22/2/2018 - Artificiële Intelligentie & Human Resources
Meetup 22/2/2018 - Artificiële Intelligentie & Human Resources
 
Re2018 Semios for Requirements
Re2018 Semios for RequirementsRe2018 Semios for Requirements
Re2018 Semios for Requirements
 
NE7012- SOCIAL NETWORK ANALYSIS
NE7012- SOCIAL NETWORK ANALYSISNE7012- SOCIAL NETWORK ANALYSIS
NE7012- SOCIAL NETWORK ANALYSIS
 
Report of Previous Project by Yifan Guo
Report of Previous Project by Yifan GuoReport of Previous Project by Yifan Guo
Report of Previous Project by Yifan Guo
 
A Review of Constraint Programming
A Review of Constraint ProgrammingA Review of Constraint Programming
A Review of Constraint Programming
 
Access To Specific Declarative Knowledge By Expert Systems The Impact Of Log...
Access To Specific Declarative Knowledge By Expert Systems  The Impact Of Log...Access To Specific Declarative Knowledge By Expert Systems  The Impact Of Log...
Access To Specific Declarative Knowledge By Expert Systems The Impact Of Log...
 
Are Human-generated Demonstrations Necessary for In-context Learning?
Are Human-generated Demonstrations Necessary for In-context Learning?Are Human-generated Demonstrations Necessary for In-context Learning?
Are Human-generated Demonstrations Necessary for In-context Learning?
 

More from butest

EL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBEEL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBE
butest
 
1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同butest
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIAL
butest
 
Timeline: The Life of Michael Jackson
Timeline: The Life of Michael JacksonTimeline: The Life of Michael Jackson
Timeline: The Life of Michael Jackson
butest
 
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
butest
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIAL
butest
 
Com 380, Summer II
Com 380, Summer IICom 380, Summer II
Com 380, Summer II
butest
 
PPT
PPTPPT
PPT
butest
 
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet JazzThe MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
butest
 
MICHAEL JACKSON.doc
MICHAEL JACKSON.docMICHAEL JACKSON.doc
MICHAEL JACKSON.doc
butest
 
Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1
butest
 
Facebook
Facebook Facebook
Facebook
butest
 
Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...
butest
 
Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...
butest
 
NEWS ANNOUNCEMENT
NEWS ANNOUNCEMENTNEWS ANNOUNCEMENT
NEWS ANNOUNCEMENT
butest
 
C-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.docC-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.doc
butest
 
MAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.docMAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.doc
butest
 
Mac OS X Guide.doc
Mac OS X Guide.docMac OS X Guide.doc
Mac OS X Guide.doc
butest
 
hier
hierhier
hier
butest
 
WEB DESIGN!
WEB DESIGN!WEB DESIGN!
WEB DESIGN!
butest
 

More from butest (20)

EL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBEEL MODELO DE NEGOCIO DE YOUTUBE
EL MODELO DE NEGOCIO DE YOUTUBE
 
1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同1. MPEG I.B.P frame之不同
1. MPEG I.B.P frame之不同
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIAL
 
Timeline: The Life of Michael Jackson
Timeline: The Life of Michael JacksonTimeline: The Life of Michael Jackson
Timeline: The Life of Michael Jackson
 
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
Popular Reading Last Updated April 1, 2010 Adams, Lorraine The ...
 
LESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIALLESSONS FROM THE MICHAEL JACKSON TRIAL
LESSONS FROM THE MICHAEL JACKSON TRIAL
 
Com 380, Summer II
Com 380, Summer IICom 380, Summer II
Com 380, Summer II
 
PPT
PPTPPT
PPT
 
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet JazzThe MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
The MYnstrel Free Press Volume 2: Economic Struggles, Meet Jazz
 
MICHAEL JACKSON.doc
MICHAEL JACKSON.docMICHAEL JACKSON.doc
MICHAEL JACKSON.doc
 
Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1Social Networks: Twitter Facebook SL - Slide 1
Social Networks: Twitter Facebook SL - Slide 1
 
Facebook
Facebook Facebook
Facebook
 
Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...Executive Summary Hare Chevrolet is a General Motors dealership ...
Executive Summary Hare Chevrolet is a General Motors dealership ...
 
Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...Welcome to the Dougherty County Public Library's Facebook and ...
Welcome to the Dougherty County Public Library's Facebook and ...
 
NEWS ANNOUNCEMENT
NEWS ANNOUNCEMENTNEWS ANNOUNCEMENT
NEWS ANNOUNCEMENT
 
C-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.docC-2100 Ultra Zoom.doc
C-2100 Ultra Zoom.doc
 
MAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.docMAC Printing on ITS Printers.doc.doc
MAC Printing on ITS Printers.doc.doc
 
Mac OS X Guide.doc
Mac OS X Guide.docMac OS X Guide.doc
Mac OS X Guide.doc
 
hier
hierhier
hier
 
WEB DESIGN!
WEB DESIGN!WEB DESIGN!
WEB DESIGN!
 

20051128.doc

  • 1. Time:11:00AM Location:New CSIE Building R110 Topic:Learning with Integer Linear Programming Inference for Constrained Output Speaker: Scott Wen-Tau Yih Abstract: In several structured classification problems, explicit and expressive constraints are crucial to enhancing the accuracy and quality of the predictions. However, it was not clear how this additional knowledge can be used in various learning frameworks. In this talk, I'll first demonstrate how constraints can be incorporated in Conditional Random Fields via a novel inference approach based on integer linear programming. Inference in CRFs and HMMs is usually done using the Viterbi algorithm, an efficient dynamic programming algorithm. In many cases, general (non-local and non-sequential) constraints may exist over the output sequence, but cannot be incorporated and exploited in a natural way by Viterbi. Our inference procedure extends CRF models to naturally and efficiently support general constraint structures. For sequential constraints, this procedure reduces to simple linear programming as the inference process. Experimental evidences of our approach will be provided in the context of an important NLP problem, semantic role labeling. One interesting phenomenon we observed in the experiments is that a simple learning plus inference scheme may outperform inference based training approaches when incorporating constraints. In the second part of my talk, I'll describe how we compared these two learning frameworks by observing their behaviors in different conditions. Experiments and theoretical justification lead to the conclusion that using inference based learning is superior when the local classifiers are difficult to learn but may require many examples before any discernible difference can be observed. Bio: Wen-tau Yih is a post-doc researcher in the Machine Learning and Applied Statistics group at Microsoft Research. He got his Ph.D. at the University of Illinois at Urbana-Champaign in May 2005. Although his current research focuses mainly on problems related to email applications and anti-spam, his research interests spread on various problems in Machine Learning and Natural Language Processing, such as learning and knowledge representation, information extraction, semantic parsing, and inference and learning for structured output. Wen-tau received both his M.S. and B.S. degrees in Computer Science from National Taiwan University. More information can be found on his homepage: http://scottyih.org/