CS 838 (Spring 2004): Statistical Relational Learning


General Course Information


Course Overview and Requirements

This is a special topics course focusing on statistical machine learning algorithms that can operate on relational data, that is, data stored in more than a single relational table.

There are three parts to this course's requirements:

  1. Students will be expected to present 2-3 papers.
  2. Students will be expected to read the paper for each class and participate in the discussions.
  3. Students will do substantial class projects of their choice. Project proposals will be due early in the semester, and a progress report will be due mid-semester. Team projects are acceptable.
Students who are auditing the course will be expected to present one paper and attend regularly.

The course schedule is at the bottom of this page.


Possible Readings

Relational Learning with Statistical Predicate Invention: Better Models for Hypertext.
Mark Craven and Sean Slattery. Machine Learning, 43(1-2): 97-119, 2001.

Prolog for First-Order Bayesian Networks: A Meta-interpreter Approach.
Hendrik Blockeel. KDD Workshop on Multi-Relational Data Mining, 2003.

Probabilistic Models for Relational Data.
D. Heckerman, C. Meek, and D. Koller. Technical Report MSR-TR-2004-30, Microsoft Research, March, 2004.

Learning Probabilistic Relational Models
Lise Getoor, Nir Friedman, Daphne Koller and Avi Pfeffer
(IJCAI'99 paper by same name and authors)

The relational vector-space model and industry classification
Abraham Bernstein, Scott Clearwater, and Foster Provost

Discriminative probabilistic models for relational data
Ben Taskar, Pieter Abbeel and Daphne Koller

Relational Reinforcement Learning
Saso Dzeroski, Luc De Raedt and Hendrik Blockeel (ICML'98 and ILP'98)

Inductive policy selection for first-order MDPs
SungWook Yoon, Alan Fern and Robert Givan (UAI'02)

A Multi-relational decision tree learning algorithm - implementation and experiments
Anna Atramentov, H. Leiva and Vasant Honavar

Speeding up multi-relational data mining
Anna Atramentov and Vasant Honavar

Categorizing unsupervised relational learning algorithms
Hannah Blau and Amy McGovern

Aggregation versus selection bias, and relational neural networks
Hendrik Blockeel and Maurice Bruynooghe

Feature extraction languages for propositionalized relational learning
Chad Cumby and Dan Roth

Individuals, relations and structures in probabilistic models
James Cussens

Operations for learning with graphical models
Wray Buntine

CLP(BN): Constraint Logic Programming for Probabilistic Knowledge
Vitor Santos Costa, David Page, Maleeha Qazi and James Cussens (UAI'03)

Learning Markov networks: Maximum bounded tree-width graphs (Symposium on Discrete Algorithms, 2001)
David Karger and Nathan Srebro

Ecosystem analysis using probabilistic relational modeling
Bruce D'Ambrosio, Eric Altendorf, and Jane Jorgensen

Dynamic probabilistic relational models (IJCAI'03)
Sumit Sanghai, Pedro Domingos and Daniel Weld

Research on statistical relational learning at the University of Washington
Pedro Domingos, Yeuhi Abe, Corin Anderson, Anhai Doan, Dieter Fox, Alon Halevy, Geoff Hulten, Henry Kautz, Tessa Lau, Lin Liao, Jayant Madhavan, Mausam, Donald J. Patterson, Matthew Richardson, Sumit Sanghai, Daniel Weld and Steve Wolfman

Relational learning for securities market regulation
Henry Goldberg

Social network relational vectors for anonymous identity matching
Shawndra Hill

Mining massive relational databases
Geoff Hulten, Pedro Domingos, and Yeuhi Abe

Representational power of probabilistic-logical models: From upgrading to downgrading
Kristian Kersting

Logical Markov decision programs
Kristian Kersting and Luc De Raedt

First-order probabilistic models for information extraction
Bhaskara Marthi, Brian Milch, and Stuart Russell

A Note on the unification of information extraction and data mining using conditional-probability, relational models
Andrew McCallum and David Jensen

The Variable precision rough set inductive logic programming model -- a statistical relational learning perspective
R. Milton, V. Maheswari and A. Siromoney

Statistical relational learning: Four claims and a survey
Jennifer Neville, Matthew Rattigan, and David Jensen

Parameter estimation for stochastic context-free graph grammars
Tim Oates, Fang Huang, and Shailesh Doshi

Aggregation and concept complexity in relational learning
Claudia Perlich and Foster Provost

Aggregation-based feature invention and relational concept classes
Claudia Perlich and Foster Provost

Relational learning problems and simple models
Foster Provost, Claudia Perlich and Sofus Macskassy

Learning probabilistic relational planning rules (Fourteenth International Conference on Automated Planning and Scheduling , 2004)
Hanna M. Pasula, Luke S. Zettlemoyer and Leslie Pack Kaelbling

Statistical relational learning for link prediction
Alexandrin Popescul and Lyle H. Ungar

A comparison of stochastic logic programs and Bayesian logic programs
Aymeric Puech and Stephen Muggleton

Principles of Learning Bayesian Logic Programs
Kristian Kersting and Luc De Raedt

Learning statistical models of time-varying relational data
Sumit Sanghai, Pedro Domingos and Daniel Weld

A new perspective of statistical modeling with PRISM
Taisuke Sato and Neng-Fu Zhou

Relational learning: A web-page classification viewpoint
Sean Slattery

Statistical modeling of graph and network data
Padhraic Smyth

Label and link prediction in relational data
Ben Taskar, Pieter Abbeel, Ming-Fai Wong, and Daphne Koller

Toward a high-performance system for symbolic and statistical modeling
Neng-Fa Zhou, Taisuke Sato, and Koiti Hasidad


Schedule of Presentations

date required readings recommended readings presenter
1/21 Learning probabilistic relational models IJCAI'99 paper by same name and authors David Page
1/23 Continuation from 1/21 -- David Page
1/26 The relational vector-space model and industry classification -- Louis Oliphant
1/28 Statistical relational learning for link prediction
Statistical relational learning for document mining
-- Mark Goadrich
2/2 Discriminative probabilistic models for relational data Label and link prediction in relational data Joe Bockhorst
2/4 A Multi-relational decision tree learning algorithm - Implementation and Experiments Speeding up multi-relational data mining Ip Kei Sam
2/6 Continuation from 2/4 -- Ip Kei Sam
2/9 Parameter estimation for stochastic context-free graph grammars -- Cheryl Barkauskas
2/13 A comparison of Stochastic logic programs and Bayesian logic programs -- Wei Luo
2/16 Constructing free energy approximations and generalized belief propagation algorithms -- Hector Corrada Bravo
2/18 Relational reinforcement learning -- Ted Wild
2/23 Inductive policy selection for first-order MDPs -- Trevor Walker
2/27 Individuals, relations and structures in probabilistic models
Operations for Learning with Graphical Models
Graphical models Keith Noto
3/1 Learning Markov networks: Maximum bounded tree-width graphs -- Cheryl Barkauskas
3/3 Principles of learning Bayesian logic programs -- Irene Ong
3/5 Continuation from 3/3 -- Irene Ong
3/22 Dynamic probabilistic relational models Learning statistical models of time-varying relational data
Research on statistical relational learning at the University of Washington
Ip Kei Sam
3/24 Categorizing unsupervised relational learning algorithms Graph-based relational concept learning
Finding frequent substructures in chemical compounds
Wei Luo
3/29 Learning probabilistic relational planning rules -- Trevor Walker
3/31 Toward a high-performance system for symbolic and statistical modeling -- Jesse Davis
4/5 Mining massive relational databases -- Yue Pan
4/7 First-order probabilistic models for information extraction Identity Uncertainty and Citation Matching (Pasula, Marthi, Milch, Russell and Shpitser)
Identity Uncertainty (Russell)
Approximate inference for first-order probabilistic languages, H. Pasula and S. Russell, IJCAI-01
Matthew Lee
4/9 Attend Ted Wild's Database Seminar at 1:30 -- Ted Wild
4/14 Continuation from 4/7 -- Matthew Lee
4/16 AI Seminar in 2310 CS at 11am -- David Musicant
4/19 Attend Eric Xing's Systems Biology Faculty Candidate talk at 10am -- Eric Xing
4/21 Continuation from 4/14 -- Matthew Lee
4/26 Project Intermediate Oral Presentations -- Registered Students
4/28 On-line Bioinformatics Seminar, 11am, 1360 Biotech Center Prediction and Design of Protein Structures and Protein-Protein Interactions David Baker
4/30 A Note on the unification of information extraction and data mining using conditional-probability, relational models -- Frank DiMaio
5/3 Prolog for First-Order Bayesian Networks: A Meta-interpreter Approach.
-- Sean McIlwain
5/5 Relational Learning with Statistical Predicate Invention: Better Models for Hypertext.
-- Mark Craven
5/7 D. Heckerman, C. Meek, and D. Koller, Probabilistic Models for Relational Data. Technical Report MSR-TR-2004-30, Microsoft Research, March, 2004 -- Maleeha Qazi