CPSC 422: Intelligent Systems
Winter Session 2009/2010 Term 2
SCHEDULE
January February March April
This schedule is subject to change. Please make sure to check it regularly, because here is where relevant material for each class (e.g., PDF lecture slides) and all the assignments will be posted. However, the textbook chapters from the additional reference book (Artificial Intelligence: Foundations of Computational Agents, by Pool and Mackworth) are only available through WebCT. Assignment due dates are provided to give you a rough sense; however, they are also subject to change. I'll post each lecture's slides right after class. I won't post additional material that I write on the slides or on the board during class. You'll have to come to class to get that). All the listed readings should be completed before class. Unless noted otherwise, all readings refer to the course textbook: Artificial Intelligence: A Modern Appropach. The additional reference, Artificial Intelligence: Foundations of Computational Agents, by Pool and Mackworth will be indicated as "P&M" when mentioned in the readings |
| Tuesday | Thursday |
5 Introduction
#1 |
7 Review of Bayesian networks
#2 |
12 Approximate Inference in Bayesian networks (background)
#3 |
14 Approximate Inference in Bayesian Networks (Algorithms)
#4 |
19 Using Bayesian Networks - Case Study: The Andes Tutoring System Slides Instructions to run and use Andes
In this class, we will discuss the following paper: Conati C., Gertner A., VanLehn K., 2002. Using Bayesian Networks to Manage Uncertainty in Student Modeling. User Modeling and User-Adapted Interaction. 12(4) p. 371-417. Make sure to have at least two questions on this reading to discuss in class. Please email your questions (with subject "questions for 422") to both conati@cs.ubc.ca, hajir@cs.ubc.ca, by 9am today. Also bring a paper copy of your questions to handin in class.
#5 |
21 Probability and time: probabilistic temporal models, algorithms
#6 |
26 Probability and time: Algorithms (cont'd), Hidden Markov Models
#7 |
28 Probability and time: PoS tagging, Dynamic Bayesian Networks, Particle Filtering
#8
|
| Tuesday | Thursday |
2 Decision theoretic planning: Intro and MDPs
#9 |
4 Decision theoretic planning: Value Iteration and Policy Iteration Assignment 1 due Friday Feb 5 Assignment 2 due Wednesday March 3
#10 |
9 Decision theoretic planning: POMDPs
To review background concepts covered in 322 see:
#11 |
11 Learning:
Introduction, Supervised Learning, Decision Trees
#12 |
Reading Week + Olympic Break |
Reading Week + Olympic Break Olympic Break |
| Tuesday | Thursday |
| 2 Lecture by Pooja Viswanathan on
applications of POMDPs in assistive technology
See http://www.cs.uwaterloo.ca/~ppoupart/software.html for code and sample problems for Symbolic Perseus algorithm for factored POMDPs
|
4 Learning: Decision Trees
|
9 Learning: Neural
Networks
#15 |
11 Midterm (click here
to see what you need to know for it)
Practice exercise on DT learning Assignment 3 due Thursday March 25,11pm
#16 |
16
Statistical learning, Learning Bayesian networks with complete
data
#17 |
18 Learning
Bayesian Networks with hidden variables
#18 |
23 Reinforcement Learning: Q-learning
Slides (cover also lecture of March 25) #19 |
25
Reinforcement Learning: Q-learning
Slides
#20 |
| 30 Reinforcement learning:
exploration/exploitation, off policy methods
due Tuesday April 13, midnight
#21 |
#22 |
| Tuesday |
Thursday |
1
Ontologies and knowledge-based systems
#22 |
|
| 6 No Class
|
8 The Semantic Web
#23 |
13 Adaptive User Interfaces (readings based
class) Make sure to have at least two questions on this reading to discuss in class. Please email your questions (with subject "questions for 422") to both conati@cs.ubc.ca, hajir@cs.ubc.ca, by 9am today.
Also bring a paper copy of your
questions to handin in class.
Assignment 4 due (@midnight) #24 |
15:
Adaptive support for interface usage (readings based
class)
Make sure to have at least one question on the reading to discuss in class. Please email your questions (with subject "questions for 422") to both conati@cs.ubc.ca, hajir@cs.ubc.ca, by 9am today,
Be prepared to discuss the Jameson's paper, which we didn't get to cover on Tu.
|
| 20 | 22 |
27 FINAL EXAM, 8:30 am , DMP 301 NOTE: you will be allowed to bring into the exam a letter sized sheet of paper with anything written on it (double sided). You may not use calculators, PDAs, robotic assistants or other electronic aids. Cristina's office hours for the period before the exam are Fri, April 23, 2--3:30 Mon, April 26, 1-2:30 or send me email if you need to make special arrangements |
1
|