Steam-powered Turing Machine University of Washington Department of Computer Science & Engineering
 CSE 574 - Artificial Intelligence II - Spring 2005
  CSE Home  About Us    Search    Contact Info 

Statistical Relational Learning

Instructor: Pedro Domingos
Office: CSE 648
Office hours: Wednesdays 2:00-2:50 p.m., and by appointment
TA: Stanley Kok
Office: CSE 216
Office hours: Mondays 4:30-5:30 p.m., and by appointment

Class meets:
Mondays, Wednesdays 3:00-4:20 p.m. in EE1 026

Lectures and Milestones

Week Topics & Lecture Notes Readings Milestones
Week 1 (Mar 28 & 30) Introduction    
Week 2 (Apr 4 & 6) Inductive logic programming Dzeroski tutorial

Project proposal due (Fri, Apr 8)

Week 3 (Apr 11 & 13) Bayesian networks Heckerman tutorial  
Week 4 (Apr 18 & 20) Markov networks Della Pietra et al.  
Week 5 (Apr 25 & 27) Markov logic networks Richardson & Domingos  
Week 6 (May 2 & 4) MLN structure learning
Probabilistic relational models
(Guest lecture by Lise Getoor)
Kok & Domingos
Getoor et al. (1)
Getoor et al. (2)
Project progress report due (Fri, May 6)
Week 7 (May 9) Discriminative training of MLNs (Parag Singla)
Relational stochastic processes (Sumit Sanghai)
Singla & Domingos
Sanghai et al.
Week 7 (May 11) Lifted inference  (Daniel Lowd)
SRL for information extraction and NLP (Michele Banko)
de Salvo Braz et al.
Emailed to class
Week 8 (May 16) Feature generation and selection in SRL (Stef Schoenmackers)
Aggregation (Pat Tressel)
Emailed to class
Perlich & Provost
Week 8 (May 18) Stochastic logic programs (Kevin Wampler)
Probabilistic models with unknown objects (Brian Ferris)
Emailed to class
Week 9 (May 23) Conditional random fields (Benson Limketkai)
Relational Markov networks (Danny Wyatt)
Lafferty et al.
Taskar et al.
Week 9 (May 25) Relational factor graphs (Lin Liao)
Max-margin Markov networks (Mike Cafarella)
Emailed to class
Taskar et al.
Week 10 (Jun 1)

1. MLNs for table induction from Web text (Michele Banko & Mike Cafarella)
2. GIS data association with MLNs (Brian Ferris)
3. Relational object maps using MLNs for mobile robot mapping (Benson Limketkai)
4. Citation prediction and suggestion using MLNs (Stef Schoenmackers)
5. Automatic inexact structured visualization (Kevin Wampler)

Week 11 (Jun 9, 2:30 pm) 1. Activity recognition using aggregate features (Lin Liao)
2. Relational mixture models (Daniel Lowd)
3. Semi-supervised and active relational learning (Sumit Sanghai)
4. Efficient inference for MLNs (Parag Singla)
5. Incorporating continuous random variables into MLNs (Pat Tressel)
6. Inferring social roles and interaction types using MLNs (Danny Wyatt)
  Project report due (Mon, Jun 6, midnight)



Students will do a project and give a seminar. Projects can be done individually or in groups of two, and are due on the last day of the last week of classes (June 3). Seminars are done individually, and are Pass/Fail; a Pass is required to complete the class. The class grade will be the project grade. An ideal project is applying statistical relational learning (SRL) to your area of research. You can also propose any other project related to SRL. Here are some ideas:


We will provide software for statistical relational learning based on Markov logic networks. Watch this space ...

Anonymous Feedback

Comments can be sent to us using this anonymous feedback form.

Course Mailing List

Instructions for joining the course mailing list.

Mailing List Archive

CSE logo Department of Computer Science & Engineering
University of Washington
Box 352350
Seattle, WA  98195-2350
(206) 543-1695 voice, (206) 543-2969 FAX
[comments to]