cross domain learning to rank with svm
play

Cross-Domain Learning-to-rank with SVM Erheng Zhong 1 1 Department of - PowerPoint PPT Presentation

Dec 09, 2023 •202 likes •620 views

Preliminary Cross-domain Learning-to-Rank Summary Cross-Domain Learning-to-rank with SVM Erheng Zhong 1 1 Department of Computer Science and Technology, HKUST COMP621U Presentation, 04/07/2011 Erheng Zhong Cross-Domain Learning-to-rank

  1. Preliminary Cross-domain Learning-to-Rank Summary Cross-Domain Learning-to-rank with SVM Erheng Zhong 1 1 Department of Computer Science and Technology, HKUST COMP621U Presentation, 04/07/2011 Erheng Zhong Cross-Domain Learning-to-rank

  2. Preliminary Cross-domain Learning-to-Rank Summary Outline Preliminary 1 Ranking Learning-to-rank Transfer Learning Cross-domain Learning-to-Rank 2 Motivations Approach: RankSVM Main Results Erheng Zhong Cross-Domain Learning-to-rank

  3. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Outline Preliminary 1 Ranking Learning-to-rank Transfer Learning Cross-domain Learning-to-Rank 2 Motivations Approach: RankSVM Main Results Erheng Zhong Cross-Domain Learning-to-rank

  4. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Definition A relationship between a set of items. A weak order or total preorder of objects. (mathematics) A central part of many information retrieval problems! Erheng Zhong Cross-Domain Learning-to-rank

  5. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Applications Search Engine Erheng Zhong Cross-Domain Learning-to-rank

  6. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Applications Recommendation System Erheng Zhong Cross-Domain Learning-to-rank

  7. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Applications Computational Advertising Erheng Zhong Cross-Domain Learning-to-rank

  8. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Outline Preliminary 1 Ranking Learning-to-rank Transfer Learning Cross-domain Learning-to-Rank 2 Motivations Approach: RankSVM Main Results Erheng Zhong Cross-Domain Learning-to-rank

  9. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Concepts Learning-to-rank [1] is to automatically construct a ranking model from training data. Training Data: Lists of <query,item> pairs with some partial order specified between pairs < X , y > ; where X = { x i = ( q k , t kj ) } ℓ i = 1 and y = { y i } ℓ i = 1 Ranking Model: A function computing relevance of items for actual queries � � = ¯ f x = ( q , t ) y Erheng Zhong Cross-Domain Learning-to-rank

  10. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Features http://research.microsoft.com/en-us/projects/ mslr/feature.aspx Erheng Zhong Cross-Domain Learning-to-rank

  11. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Framework Erheng Zhong Cross-Domain Learning-to-rank

  12. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Approaches Three groups with different input representations and loss functions: Pointwise Approach: Each query-document pair in the training data has a numerical or ordinal score. A regression problem. Pairwise Approach: A binary classifier which can tell which document is better in a given pair of documents. The goal is to minimize average number of inversions in ranking. Listwise Approach: They directly optimize the value of one evaluation measure. Erheng Zhong Cross-Domain Learning-to-rank

  13. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Approaches Three groups with different input representations and loss functions: Pointwise Approach: Each query-document pair in the training data has a numerical or ordinal score. A regression problem. Pairwise Approach: A binary classifier which can tell which document is better in a given pair of documents. The goal is to minimize average number of inversions in ranking. Listwise Approach: They directly optimize the value of one evaluation measure. Erheng Zhong Cross-Domain Learning-to-rank

  14. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Approaches Three groups with different input representations and loss functions: Pointwise Approach: Each query-document pair in the training data has a numerical or ordinal score. A regression problem. Pairwise Approach: A binary classifier which can tell which document is better in a given pair of documents. The goal is to minimize average number of inversions in ranking. Listwise Approach: They directly optimize the value of one evaluation measure. Erheng Zhong Cross-Domain Learning-to-rank

  15. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Approaches Three groups with different input representations and loss functions: Pointwise Approach: Each query-document pair in the training data has a numerical or ordinal score. A regression problem. Pairwise Approach: A binary classifier which can tell which document is better in a given pair of documents. The goal is to minimize average number of inversions in ranking. Listwise Approach: They directly optimize the value of one evaluation measure. Erheng Zhong Cross-Domain Learning-to-rank

  16. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Outline Preliminary 1 Ranking Learning-to-rank Transfer Learning Cross-domain Learning-to-Rank 2 Motivations Approach: RankSVM Main Results Erheng Zhong Cross-Domain Learning-to-rank

  17. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Concepts and Notations Transfer learning [2] refers to the machine learning framework in which one extracts knowledge from some auxiliary domains to help boost the learning performance in a target domain. Auxiliary domain: D s = { X s , y s } Target domain: D t = { X ℓ , y ℓ ; X u } P s (( x ) , y ) � = P t (( x ) , y ) Erheng Zhong Cross-Domain Learning-to-rank

  18. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Approaches “ what to transfer ” [2] Model-based Transfer: Discover shared parameters or prior between cross-domain models. Feature-based Transfer: Find a “good” feature representation that reduces the difference and prediction error between domains. Instance-based Transfer: Re-weight some labeled data in the auxiliary domain for use in the target domain. Erheng Zhong Cross-Domain Learning-to-rank

  19. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Approaches “ what to transfer ” [2] Model-based Transfer: Discover shared parameters or prior between cross-domain models. Feature-based Transfer: Find a “good” feature representation that reduces the difference and prediction error between domains. Instance-based Transfer: Re-weight some labeled data in the auxiliary domain for use in the target domain. Erheng Zhong Cross-Domain Learning-to-rank

  20. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Approaches “ what to transfer ” [2] Model-based Transfer: Discover shared parameters or prior between cross-domain models. Feature-based Transfer: Find a “good” feature representation that reduces the difference and prediction error between domains. Instance-based Transfer: Re-weight some labeled data in the auxiliary domain for use in the target domain. Erheng Zhong Cross-Domain Learning-to-rank

  21. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Approaches “ what to transfer ” [2] Model-based Transfer: Discover shared parameters or prior between cross-domain models. Feature-based Transfer: Find a “good” feature representation that reduces the difference and prediction error between domains. Instance-based Transfer: Re-weight some labeled data in the auxiliary domain for use in the target domain. Erheng Zhong Cross-Domain Learning-to-rank

  22. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Applications Text classification Sentiment analysis Image classification Name-entity recognition WiFi localization Spam Filtering . . . Ranking! Erheng Zhong Cross-Domain Learning-to-rank

  23. Preliminary Ranking Cross-domain Learning-to-Rank Learning-to-rank Summary Transfer Learning Applications Text classification Sentiment analysis Image classification Name-entity recognition WiFi localization Spam Filtering . . . Ranking! Erheng Zhong Cross-Domain Learning-to-rank

  24. Preliminary Motivations Cross-domain Learning-to-Rank Approach: RankSVM Summary Main Results Outline Preliminary 1 Ranking Learning-to-rank Transfer Learning Cross-domain Learning-to-Rank 2 Motivations Approach: RankSVM Main Results Erheng Zhong Cross-Domain Learning-to-rank

  25. Preliminary Motivations Cross-domain Learning-to-Rank Approach: RankSVM Summary Main Results Sparsity Problem No enough labeled data in the current domain. Heterogeneous feature spaces? Text search ⇒ Image search? Out-of-date data? Log data past years ⇒ Search task this year? Heterogeneous tasks? Web page ranking ⇒ Expert finding? . . . Erheng Zhong Cross-Domain Learning-to-rank

Recommend

Kicking Down the Cross Domain Door Techniques for Cross Domain Exploitation Billy K Rios (BK) and

Kicking Down the Cross Domain Door Techniques for Cross Domain Exploitation Billy K Rios (BK) and

Kicking Down the Cross Domain Door Techniques for Cross Domain Exploitation Billy K Rios (BK) and Raghav Dube Tabbed Browsing Rich Content Mash-ups Cookies JSON Ajax Implication of Cross Domain Attacks Implication of Cross Domain Attacks

580 views • 31 slides
10. Learning to Rank Outline 10.1. Why Learning to Rank (LeToR)? 10.2. Pointwise, Pairwise,

10. Learning to Rank Outline 10.1. Why Learning to Rank (LeToR)? 10.2. Pointwise, Pairwise,

10. Learning to Rank Outline 10.1. Why Learning to Rank (LeToR)? 10.2. Pointwise, Pairwise, Listwise 10.3. Gathering User Input 10.4. LeToR Evaluation 10.5. Beyond Search Advanced Topics in Information Retrieval / Learning to Rank 2 10.1.

1.01k views • 34 slides
2 3 4 5 8 9 MINNEAPOLIS MILWAUKEE MSA RANK #16 MSA RANK #39 CHICAGO MSA RANK #3

2 3 4 5 8 9 MINNEAPOLIS MILWAUKEE MSA RANK #16 MSA RANK #39 CHICAGO MSA RANK #3

2 3 4 5 8 9 MINNEAPOLIS MILWAUKEE MSA RANK #16 MSA RANK #39 CHICAGO MSA RANK #3 INDIANAPOLIS MSA RANK #34 DETROIT MSA RANK #14 COLUMBUS MSA RANK #33 BALTIMORE/ DC CORRIDOR MSA RANK #21 CINCINNATI MSA RANK #28 ATLANTA MSA RANK

726 views • 44 slides
CoNet: Collaborative Cross Networks for Cross-Domain Recommendation Guangneng Hu*, Yu Zhang, and

CoNet: Collaborative Cross Networks for Cross-Domain Recommendation Guangneng Hu*, Yu Zhang, and

CoNet: Collaborative Cross Networks for Cross-Domain Recommendation Guangneng Hu*, Yu Zhang, and Qiang Yang CIKM 2018 Oct 22-26 (Mo-Fr), Turin, Italy 1 Recommendations Are Ubiquitous: Products, Medias, Entertainment Amazon 300

277 views • 27 slides
Learning to Rank Learning to Rank with Partially-Labeled Data with Partially-Labeled Data Kevin

Learning to Rank Learning to Rank with Partially-Labeled Data with Partially-Labeled Data Kevin

Learning to Rank Learning to Rank with Partially-Labeled Data with Partially-Labeled Data Kevin Duh University of Washington (Joint work with Katrin Kirchhoff) 1 Motivation Motivation Machine learning can be an effective solution for

472 views • 29 slides
Domain adaptation model for retinopathy detection from cross-domain OCT images Jing Wang 1;2 ,

Domain adaptation model for retinopathy detection from cross-domain OCT images Jing Wang 1;2 ,

Domain adaptation model for retinopathy detection from cross-domain OCT images Jing Wang 1;2 , Yiwei Chen 2 , Wanyue Li1; 2 , Wen Kong 1;2 , Yi He 2 , Chuihui Jiang *3 , Guohua Shi *2;4 1 University of Science and technology of China, Hefei 230026,

348 views • 9 slides
Cross-Layer and Cross-Domain QoS Signalling Using BGP 8th Wrzburg Workshop on IP: Joint

Cross-Layer and Cross-Domain QoS Signalling Using BGP 8th Wrzburg Workshop on IP: Joint

Cross-Layer and Cross-Domain QoS Signalling Using BGP 8th Wrzburg Workshop on IP: Joint EuroNF, ITC, and ITG Workshop on &quot;Visions of Future Generation Networks&quot; (EuroView2008) Thomas Martin Knoll Chemnitz University of Technology

101 views • 7 slides
Introduction to Information Retrieval http://informationretrieval.org IIR 15-2: Learning to Rank

Introduction to Information Retrieval http://informationretrieval.org IIR 15-2: Learning to Rank

Machine-learned relevance Learning to rank Introduction to Information Retrieval http://informationretrieval.org IIR 15-2: Learning to Rank Hinrich Sch utze Institute for Natural Language Processing, Universit at Stuttgart 2011-08-29

1.3k views • 115 slides
Efficient Domain Generalization via Common-Specific Low-Rank Decomposition * Sunita Sarawagi 1

Efficient Domain Generalization via Common-Specific Low-Rank Decomposition * Sunita Sarawagi 1

Efficient Domain Generalization via Common-Specific Low-Rank Decomposition * Sunita Sarawagi 1 Vihari Piratla 12 Praneeth Netrapalli 2 1 Indian Institute of Technology, Bombay 2 Microsoft Research, India * ICML 2020,

531 views • 28 slides
Learning to Rank with Learning to Rank with Partially-Labeled Data Partially-Labeled Data Kevin

Learning to Rank with Learning to Rank with Partially-Labeled Data Partially-Labeled Data Kevin

Learning to Rank with Learning to Rank with Partially-Labeled Data Partially-Labeled Data Kevin Duh University of Washington 1 The Ranking Problem The Ranking Problem Definition: Given a set of objects, sort them by preference. objectA

716 views • 59 slides
Online Learning to Rank with Features Authors: Shuai Li, Tor Lattimore, Csaba Szepesvri The

Online Learning to Rank with Features Authors: Shuai Li, Tor Lattimore, Csaba Szepesvri The

Online Learning to Rank with Features Authors: Shuai Li, Tor Lattimore, Csaba Szepesvri The Chinese University of Hong Kong DeepMind University of Alberta Learning to Rank Amazon, YouTube, Facebook, Netflix, Taobao 1 Online Learning to

530 views • 17 slides
Requirements for Policies in Cross-Domain Service

Requirements for Policies in Cross-Domain Service

Requirements for Policies in Cross-Domain Service Composi8on Ulrich Pinsdorf, Microso&gt; Jan Schallaboeck, ULD Stuart Short, SAP

361 views • 7 slides
Learning to rank search results Voting algorithms, rank combination methods Web Search Andr

Learning to rank search results Voting algorithms, rank combination methods Web Search Andr

Learning to rank search results Voting algorithms, rank combination methods Web Search Andr Mouro, Joo Magalhes 1 2 How can we merge these results? Which model should we select for our production system? Not trivial. Would

616 views • 40 slides
Learning Goals 1 Practice Questions 1 3 2 The Holmes scenario 2 1 Learning Goals 1

Learning Goals 1 Practice Questions 1 3 2 The Holmes scenario 2 1 Learning Goals 1

CS 486/686 Probabilities by Xinda Li and Alice Gao 2 Given a description of a domain or a probabilistic model for the domain, determine whether two variables are unconditionally independent. Given a description of a domain or a

316 views • 13 slides
Strong Baselines for Neural Semi-supervised Learning under Domain Shift Sebastian Ruder Barbara

Strong Baselines for Neural Semi-supervised Learning under Domain Shift Sebastian Ruder Barbara

Strong Baselines for Neural Semi-supervised Learning under Domain Shift Sebastian Ruder Barbara Plank Learning under Domain Shift 2 Learning under Domain Shift State-of-the-art domain adaptation approaches 2 Learning under Domain Shift

1.37k views • 112 slides
The Cross- -domain Information domain Information The Cross Exchange Framework (CIEF) Exchange

The Cross- -domain Information domain Information The Cross Exchange Framework (CIEF) Exchange

The Cross- -domain Information domain Information The Cross Exchange Framework (CIEF) Exchange Framework (CIEF) 20 May 2008 20 May 2008 Paul Shaw, SPAWARSYSCOM Paul Shaw, SPAWARSYSCOM Dr. David J. Roberts, iBASEt, Inc. Dr. David J.

856 views • 59 slides
Controlling Fairness and Bias in Dynamic Learning-to-Rank ACM SIGIR 2020 Marco Morik * ,

Controlling Fairness and Bias in Dynamic Learning-to-Rank ACM SIGIR 2020 Marco Morik * ,

Controlling Fairness and Bias in Dynamic Learning-to-Rank ACM SIGIR 2020 Marco Morik * , Ashudeep Singh * , Jessica Hong , Thorsten Joachims TU Berlin, Cornell University Dynamic Learning-to-Rank +1 +1 +1 +1 +1 1 2

224 views • 20 slides
Learning to Rank Weinan Zhang Shanghai Jiao Tong University http://wnzhang.net

Learning to Rank Weinan Zhang Shanghai Jiao Tong University http://wnzhang.net

2019 EE448, Big Data Mining, Lecture 9 Learning to Rank Weinan Zhang Shanghai Jiao Tong University http://wnzhang.net http://wnzhang.net/teaching/ee448/index.html Content of This Course Another ML problem: ranking Learning to rank

1.11k views • 30 slides
1 SVD applications: rank, column, row, and null spaces Rank : the rank of a matrix is equal to:

1 SVD applications: rank, column, row, and null spaces Rank : the rank of a matrix is equal to:

Statistical Modeling and Analysis of Neural Data (NEU 560) Princeton University, Spring 2018 Jonathan Pillow Lecture 3A notes: SVD and Linear Systems 1 SVD applications: rank, column, row, and null spaces Rank : the rank of a matrix is equal to:

462 views • 3 slides
Learning Learning to Rank Social Media Liebling mit ber 360.000 Facebook Fans Mehrfach

Learning Learning to Rank Social Media Liebling mit ber 360.000 Facebook Fans Mehrfach

Learning Learning to Rank Social Media Liebling mit ber 360.000 Facebook Fans Mehrfach bester Arbeitgeber Deutschlands im Handel und Konsum laut Arbeitgeber-Ranking Fabian Klenk Product Owner Search shopping24 internet group

563 views • 29 slides
Ac#ve Learning and Search on Low-Rank Matrices Dougal J.

Ac#ve Learning and Search on Low-Rank Matrices Dougal J.

Ac#ve Learning and Search on Low-Rank Matrices Dougal J. Sutherland with Barnabs Pczos and Jeff Schneider Collabora#ve predic#on NeHlix problem:

600 views • 24 slides
A Hierarchical Framework for Cross Domain MapReduce Execution Yuan Luo 1 , Zhenhua Guo 1 ,

A Hierarchical Framework for Cross Domain MapReduce Execution Yuan Luo 1 , Zhenhua Guo 1 ,

A Hierarchical Framework for Cross Domain MapReduce Execution Yuan Luo 1 , Zhenhua Guo 1 , Yiming Sun 1 , Beth Plale 1 , Judy Qiu 1 , Wilfred W. Li 2 1 School of Informatics and Computing, Indiana University 2 San Diego Supercomputer Center,

356 views • 20 slides
Identifying Transferable Information Across Domains for Cross-domain Sentiment Classification

Identifying Transferable Information Across Domains for Cross-domain Sentiment Classification

Identifying Transferable Information Across Domains for Cross-domain Sentiment Classification Authors: Raksha Sharma, Pushpak Bhattacharyya, Sandipan Dandapat and Himanshu Sharad Bhatt Affiliation: IIT Bombay &amp; Xerox Research Center of

125 views • 10 slides
Towards a Reference Architecture for Service- Oriented Cross Domain Security Infrastructures Wen

Towards a Reference Architecture for Service- Oriented Cross Domain Security Infrastructures Wen

Towards a Reference Architecture for Service- Oriented Cross Domain Security Infrastructures Wen Zhu Dr. Lowell Vizenor Dr. Avinash Srinivasan 7 th International Conference on Internet and Distributed Computing Systems Agenda Background

128 views • 12 slides

More recommend