[an error occurred while processing this directive]

ICS-175A, Bayesian and Constraint Networks, Spring 2003


  • Classroom: IERF B015
  • Days: Tuesday & Thursday
  • Time: 12:30 - 1:50pm
  • Instructor: Rina Dechter - dechter@ics.uci.edu
    Office: CS 424-E
    Hours: Thursday, 11:00 am - 12:00 pm.
  • TA: Robert Mateescu
    Office: CS/E 332
    Hours: Monday, 11:00 am - 12:30 pm.
  • Reader: Radu Marinescu, radum@ics.uci.edu
    Office : Office: CS/E 331
    Hours: Monday, 11:00 am - 1:00 pm.
Course Goals:
Requirements
Students are required to do a project in Artificial Intelligence. Weekly progress reports will be graded.

Final project: submit a report + code + demo + presentation.

Students will be required to work independently and be expected to acquire all the knowledge necessary for the project. They will have to fill-up the necessary gaps in their background. TA and instructor will help with an introductory overview and refer the students to the appropriate literature. In particular, basic knowledge in Bayesain networks and Constraint processing will be necessary.

We will have a class meeting each week on Tuesday. we will have individual/group meetings and TA meetings each Thursday.

Grading:
Final grade: Weekly project reports, 20%.
Demo-presentation: 30%
Final report: 50%.
Projects Ideas:
There will be two types of projects. Project of building an AI system that provide advise in some area. We will use graphical model frameworks and focus primarily on Bayesian Networks (BN). Students can choose projects from other areas in AI, such as search and constraint satisfaction, and planning. The second type is "Research projects". Students will delve into a research question with a graduate student and will conduct empirical investigation pursuing the question at stake.

Students can select a proposed prject or may also come up with a proposal of their own which is relevant to an AI class.

System Building Projects:
  1. Primary focus of the lab:

    The project is to build a Bayesian network that models a domain and makes some inferences. Available tools such as REES, Hugin and JavaBayes can be used. The system can be built using knowledge acquisition from expert in the domain or by learning from data or both. Following are domains used in the past.

    Admission to a Phd program.
    Loan expert advisor
    Basketball simulator
    Handicappers (piking winners in horseracing)
    Selfpreserving building

    Choose your domain, model it and deonstrate query processing over your model by REES/JavaBayes/Hugin.

  2. Some Ideas For Domains
  3. Requirements For the Bayesian Modeling Project
  4. Systems based on constraint networks

    TA assignments: Given class schedules for quarter, the number of TA needed for each class, TA’s preferences and qualification, and instructors’ choice of preferred TA, schedule the TA’s in a way that maximize some measure of staisfaction. (Real data may be available)(Talk to Andre Deloach)

    Class scheduling: The problem is to find a schedule for classes, classrooms, and teachers for a teaching seting (e.g., a high school, a computer science department). Measure of satisfaction can be to minimize the number of weighted constraints violated. Students can use the REES tool to model and run their algorithms.

    Combinatorial auctions:

Research with a graduate student:
  1. Triangulation algorithms, induced-width, cycle-custet,w-cutste.
    Many algorithms applied to graphical models (Bayesian networks and constraint-networks) have complexity related to a graph parameter known as induced-width. The task will be to implement a variety of approximation methods (greedy methods, local search methods) for induced-width and compare on real benchmarks and randomly generated networks.

    The problem of findinging minimum induced-width is related to graphs' preprocessing for triangulation. The problem is to find atriangulation of a graph such that the maximum size of its cliques is minimal. In a recent paper several rules are used to preprocess the initial graph in orderto reduce it to a smaller graph before triangulating it. The triangulationof the original graph can then be obtained by reversing the reductionsteps.

    Another graph related investigations are to find a cycle-sutest of a graph. Yet another related problem is to find a w-cutset of a graph.

    References

    1. Hans L. Bodlaender et al., "Pre-processing for Triangulation of Probabilistic Networks", Proceedings of UAI, 2001
    2. Judea Pearl. "Probabilistic Reasoning in Intelligent Systems". Morgan Kaufman, ch. 3.2.4 (graph triangulation algorithm)
    3. A. Becker, R. Bar-Yehuda, D. Geiger, "Random Algorithms for the Loop Cutset Problem", UAI, 1999
    4. A. Becker, D. Geiger, "Approximation Algorithms for the Loop Cutset Problem", UAI 1994

    Most of these papers can be retrieved from http://citeseer.nj.nec.com.

  2. Experimenting with Iterative belief propagation.
  3. Bozhena's Debuging domain.

Resources on the Internet

Schedule:

Week Topic Date  
Week 1
  • Overview of necessary background in Bayes networks. Start forming groups for projects.
04-01
Week 2
  • Presentation of specific projects. Each group provides a proposal for two possible projects it considers.
04-08
Week 3
  • Progress report.
04-15
Week 4
  • Progress report.
04-22
Week 5
  • Progress report.
04-29
Week 6
  • Mid-quarter progress report and presentation.
05-06
Week 7
  • Progress report.
05-13
Week 8
  • End of eight week draft of final report.
05-20
Week 9
  • Demo-presentations.
05-27
Week 10
  • Demo-presentations.
06-05
Week 11
  • (Finals): final report + code + demo-presentation.
06-12