CS 420, Fall 2011



Syllabus: Fall 2011

Reading responses are due by 6am on the date listed below. For instance, the reading response for 9/2 (which is very short) is due at 6am on 9/2 (on Moodle).
Assignments are due by noon the day they are listed beow. For instance, Project 0 is due by noon on 9/5.

Many of the slides used in class are based on Berkeley's AI class taught by Dan Klein and many projects are based on the Pac-Man projects by John DeNero and Dan Klein.



Schedule
Date
Topic
Reading
Assignment
8/29 Gonna rock you like a hurricane!    
8/31 Introduction none  
9/2 Introduction/GOFAI Read chapter 1. You can skip/skim 1.2 and 1.3.
Post to moodle 9/1 forum.
Register for the free Stanford AI class
(we may use it as a resource later in the semester).
9/5 Classic Search Chap. 2
For your response, select one real or imaginable agent not discussed in class or in the readings, give a
PEAS description of the task environment, and characterize it in terms of the properties listed in Section 2.3.2.
Project 0: Tutorial (due by noon)
9/7 Classic Search: DFS, BFS, UCS Chap. 3-3.4
In your response, include a solution to at least one exercise from the end of the chapter.
 
9/9 Classic Search: Intro. to Heuristics Rest of Chap. 3
In your response, include a solution to at least one exercise from the end of the chapter.
 
9/12 Classic Search: More on A* none  
9/14 Beyond Classic Search/Genetic Algorithms Chap. 4-4.4  
9/16 More Genetic Algorithms Rest of Chap. 4, Sections 1--3 of the NEAT paper (you may want to skim the rest).  
9/19 Adversarial Search Read all of Chapter 5. In the response, include the solution to at least on problem  
9/21 Adversarial Search (same slides as above) none  
9/23 Adversarial Search none  
9/24     Project 1, Search: due midnight (00:01 on 9/25 is late)
9/26 Probabilities and Expectations (Review?) Chapter 13 through end of 13.5 (13.6 is also interesting (Wumpus World is defined in Section 7.2))  
9/28 Utilities Chapter 14-14.2  
9/30 Utilities and UCT Read pages 1-4 of this paper. No response is required but we will discuss the paper in class.  
10/3 Utilities and MDPs 16-16.3  
10/5 Project 1 17-17.3  
10/7 Utilities and MDPs none  
10/8     Project 2, Adversarial Search: due midnight (00:01 on 8/9 is late)
10/12 Reinforcement Learning 21-21.3  
10/14 Reinforcement Learning none  
10/17 Reinforcement Learning Rest of chapter 21  
10/19 Reinforcement Learning    
10/21
Midterm
   
10/24 Review midterm    
10/26 Probabilistic Reasoning 14-14.4.1 (skim 14.4)  
10/28 Probabilistic Reasoning    
10/29     Project 3, RL: due midnight (00:01 on 10/30 is late)
10/31 Probabilistic Reasoning/Bayesian Networks 14.5 (skim 14.7)  
11/2 Bayesian Networks    
11/4 Bayesian Networks: Join and Sum Out    
11/7 Bayesian Networks    
11/9 Bayesian Networks and Actions Pratice Problem    
11/11 Probabilistic Reasoning over Time 15.5, 15.6 (15.4 optional)  
11/14 Particle Filters Skim 15-15.3 (reading response not required)  
11/16 DBNs/Classification    
11/18 Perceptrons/Neural Networks + Slides from Mitchell 18-18.2, 18.7 (no reading response)  
11/28 Guest Lecture   Project 4, Tracking: due at 11:59pm
11/30 Classical Planning Respond to chapter 10-10.3 (Skim rest of chapter)  
12/2 Classical Planning    
12/5 Guest Lecture Respond to chapters 26 and 27  
12/7 Ethics/Philosophy    
12/9 Wrap-up   Extra Optional Project: due at 11:59pm (but not counted as late until 12/14 at 11:59pm)
12/14
Final: 8-11am
Optional papers: 1 2
Potentially useful for studying: 2, 3, 4, 5 1, 2, 5, 6, 7 1, 2, 4, 5, 6 2, 3, 4, 5 (Note: Some have solutions on webpage.)
Additional problems: 15.15, 15.16, 10.2, 10.3, 10.4, 10.5, 10.6