CSE 327 AI Theory and Practice, Spring 2016
Professor Jeff Heflin
TTh 1:10-2:25pm, Sinclair Auditorium
Check here for updates regarding the course.
- 5/13/16 - I discovered an error on Slide #6 of ch18-learning.pptx. After example 2 of epoch 2, the change to W0 should be 0.1, and thus W0 for examples 3 and 4 is 0.3 (not 0.1). This also changes the in column for both examples, but does not change the out or Err. The corrected slide is now posted.
- 5/11/16 - HW #7 has been graded. You can pick up your assignment during Thursday office hours (10-11:30am) or any time I am in my office this week. The solutions are now online.
- 5/9/16 - I will hold my usual office hours this week (on May 9 and May 12). After that, office hours are by appointment only.
- 5/5/16 - The final study guide is posted here.
- 3/8/16 - Tonight at 11pm, Google's AlphaGo program will play the top human Go player. There will also be matches on March 9th, 11th, 12th, and 14th. Watch the matches here: https://www.youtube.com/channel/UCP7jMXSY2xbc3KCAE0MHQ-A
- 2/14/16 - Due to travel to the national AI conference, class is canceled on Tue., Feb. 16. Also, I will not hold office hours on Monday, Feb. 15.
- 2/8/16 - In Homework #1, exercise #4, the SUCCESSOR-FN is not defined. Assume SUCCESSOR-FN(s) returns a set of <a,s'> pairs, where a is an action that can be performed in s and s' is a state that results if a is performed in s. Also, LEGAL-ACTIONS is just the ACTIONS function as described on p. 67.
This course will provide a general introduction to Artificial Intelligence(AI). We will discuss what AI is, survey some of the major results in the field, and look at a few promising directions. In particular, we will seek answers to questions such as:
Our examination of these problems will focus on various data structures and algorithms that have been proposed as solutions.
- How do you represent and reason with general-purpose knowledge?
- How can a robot or artificial agent formulate a plan to achieve a task?
- How can an agent make good decisions given uncertainty about its environment?
- How can an agent learn in order to improve its behavior or cope with unanticipated situations?
For details about course content, grading, and assignments, see the class syllabus.
Mon. 1-2:30pm, Thr. 10-11:30am and by appointment in Packard Lab 330
Each of the homeworks will be made available here after they are
handed out in class. The online versions of the homework are in PDF format.
Your readings will be listed below as they are assigned. Unless otherwise specified, all readings are from our textbook, Artificial Intelligence: A Modern Approach.
|Read Ch. 1-2.3 (pp. 1-46)||1/28|
|Read Sect. 2.4-2.5 (pp. 46-59)||2/2|
|Read Sect. 3.1-3.4 (pp. 64-91)||2/4|
|Read Sect. 3.5-3.7 (pp. 92-109)||2/9|
|Read Sect. 5.1-5.2, 5.4, 5.7-5.9 (pp. 161-167, 171-176, 180-186)||2/18|
|Read Sect. 7.1-7.4 (pp. 234-249)||2/23|
|Read Sect. 7.5.3-7.5.4, 7.7-7.8 (pp. 256-259, 265-275)||2/25|
|Read Sect. 8.1-8.2 (pp. 285-300)||3/1|
|Read Sect. 8.3-8.5 (pp. 300-313)||3/3|
|Read An Introduction to Prolog Programming by Ulle Endriss, Chapter 1 (pp. 1-12) and Sect. 6.1 (pp. 61-63)|
Read Sect. 9.1-9.2 (pp. 322-329)
|Read Sect. 9.4, 12.1-12.2 (pp. 337-345, 437-445)||3/24|
|Read Sect. 12.5, 12.7-12.8 (pp. 453-458, 462-468)|
Read Sect. 10.1 (pp. 366-372)
|Read Sect. 10.2.1-10.2.3 (pp. 373-379)||3/31|
|Read Sect. 10.3 (pp. 379-387)|
|Read Sect. 13.1-13.2 (pp. 480-490)||4/7|
|Read Sect. 13.3-13.7 (pp. 490-503)||4/12|
|Read Sect. 14.1-14.2 (pp. 510-518)||4/14|
|Read Sect. 14.4 (pp. 522-530)||4/19|
|Read Sect. 16.1-16.3 (pp. 610-621)||4/21|
|Read Sect. 18.1-18.3 (pp. 693-707)||4/26|
|Read Sect. 18.4, 18.6 (pp. 708-713,717-727)||4/28|
|Read Sect. 18.7 (pp. 727-737)||5/3|
|Read Sect. 18.8-18.9 (pp. 737-748)||5/5|
Additional Class Materials
- Contains information on course content, grading, assignments, and office
- Supplemental Slides
- This directory contains the slides that I use in class. Note, these slides only cover part of the lecture, and should not be used as a substitute for it.
- Search Strategy Code
- A ZIP file containing Java classes that implement three different best-first search strategies. The code is designed to be extended with definitions of specific search problems, so that it can then be used to solve those problems. This code should be used when performing the Extra Credit exercise of HW #2. This code is intended only for use in conjunction with CSE 327 at Lehigh, and is not authorized for any other purpose.
- An Introduction to Prolog Programming by Ulle Endriss
- Gives a light weight introduction to Prolog syntax, queries, and style.
- Midterm Study Guide
- This document briefly discusses the format of the test, and provides a partial list of topics you need to know for the test. It also explicitly lists topics You do not need to know.
- SWI-Prolog is free software. If you are using a personal machine, you can download SWI-Prolog from the web page listed below. If you are in a Lehigh computer lab, you should install it via the Lehigh Software page. A link to the online reference manual is also provided.
- Sample Prolog programs
- These are the examples that were shown in class
- Final Study Guide
- This document briefly discusses the format of the final, and provides a partial list of topics you need to know for the test. It also explicitly lists topics you do not need to know.