brain

CSC480 ARTIFICIAL

INTELLIGENCE

psyche

Dr. Susan Imberman Office Hours: Monday 5:00PM – 6:30PM 1N208

Wednesday 5:00PM – 6:30PM 1N208

Professor Imberman's email: imberman at mail dot csi dot cuny dot edu (NO SPACES!! The space is there to thwart any Spam crawlers. Replace the word at with @ and dot with . )

SYLLABUS

LINK to Norvig and Russell’s PowerPoint Slides

Blackboard URL is: https://doorway.cuny.edu/Doorway?ApplicationName=blackboard

Blackboard works best with Mozilla Firefox

January | February | March | April | May

Date

Topic and Assignments

1/27

What is AI? The Turing TEST

Assignment 1 – A Modified Turing Test

1/29

A History of AI

A Brief History of AI

A VERY Brief History of AI – the paper

2/4

Online – Class work on History of AI WIKI due

I will be in class and available for questions since this is the first online class this semester.

2/6

Intelligent Agents For Video Click Here

WIKI is fixed!! Don’t forget to take the AI History test in the Assignments folder of Blackboard

Lecture Notes

2/11

Intelligent Agents cont. Uninformed Search For Video Click Here

Lecture Notes

How to evaluate a paper!!

Part A due (Assignment 1 – A Modified Turing Test)

2/13

Blackboard Brief History of AI Wiki due

AI History test in the Assignments folder of Blackboard due today

Constructing our robots - http://www.cs.csi.cuny.edu/~imberman/ai/Build%20Gear%20Box.htm

2/18

No Class – Presidents day

2/20

Uninformed Search continued For Video Click Here

Lecture Notes

2/25

Uninformed Search continued For Video Click Here Those who were not in class need to hand in their answers to the quiz questions embedded in the video. You can hand in the screen capture of the question or copy the question and hand it in. Place the quiz answers into your digital drop box in blackboard.

Lecture Notes

Assignment – Turing Test Part 2: In the Assignment section of our Blackboard site, take the Turing test survey. Due 3/3

Make sure you complete the AI History test in the Assignments section

Using one of the uninformed search methods discussed in class:

a) write a program, in the language of your choice that finds a solution to the cannibals and missionaries problem. To get full credit, you must demonstrate that your program works. Hand in the source code and your solution.

b) Trace a depth first solution for cannibals and missionaries by hand. Hand in the entire search tree, with your solution

A nice description of the problem can be found on pg 90, problem 3.8 of your text.

Extra Credit (Worth 1 percentage point added to your grade)

Implement both breadth first and depth first search to solve the cannibals and missionaries problem in the programming language of your choice. (Note this meets the requirements of Assignment 3 as well.) Run experiments that compare both methods with respect to time and memory used. Write a short "research paper" describing the problem, your methods, results, a discussion of your results, and a conclusion.

******** DUE 3/17**********

2/27

Robot Lab

3/3

Uninformed Search part 2, Informed search For Video Click Here

Lecture Notes

Survey’s must be completed by today

Group HW can be done individually or in a group. No more than 3 people in a group. If the HW is done in a group then only one HW set is to be handed in for the entire group. Everyone in the group will receive the same grade.

pg 56 - 57 2.1, 2.4, pg 89 - 90 3.1, 3.7 a, d, 3.8, SAMPLE MIDTERM EXAM - CLICK HERE

************DUE 3/10*************

3/5

Informed Search

3/10

Group HW due

Answers Sample Midterm For Video Click Here

Answers Sample Midterm2

History of AI document, a copy of the Blackboard assignment

3/12

Midterm Exam - SAMPLE MIDTERM EXAM - CLICK HERE

3/17

Cannibals and Missionary programs due

Heuristics For Video Click Here

3/19

Robot Lab

Assignment: Solving the 16 Puzzle Due 3/31

3/24

No Class

3/26

Class Cancelled

Work on:

Cannibals and Missionaries

16 puzzle Assignment

Robot Programs For Interactive C go to www.kipr.org

CSC 126 Robot Lab

3/31

Cannibals and Missionaries due today

16 puzzle due today

Genetic Algorithms For Video Click Here

4/2

Robot Lab – Make sure you read the soccer agent instructions.

Make sure I initialize your checkoff sheet

4/7

What is Machine Learning? For Video part 1 Click Here For Video Part 2 Click Here

Lecture notes

4/9

Neural Networks Lecture Notes part 1 Lecture Notes part2 are in the document section of blackboard

4/14

Neural Networks For Video Click Here

Lecture Notes

Decision Tree and Neural Network Slides from Tom Mitchell

Neural Nets and Decision Tree Handout Do problems 2 and 3 to hand in. These were from old final exams!!!

Code for perceptron

Code to run trained perceptron

Code for sigmoid unit

Code to run trained sigmoid unit

Neural Network Program Due 4/30 NO EXCEPTIONS!! Modify the generation5 code to work with your robot http://www.generation5.org/content/2000/cbpnet.asp

4/16

Robot Lab Read the soccer agent instructions . I have added additional instructions

Work on your robots over spring break. You should have it programmed to do the neural network learning by 4/30

All three programs are to be handed in on 4/30

4/19 – 4/27

Spring Break

4/28

Decision Trees For Video Click Here

Lecture Notes - Entropy and Information Gain

Decision Tree Calculations for Play Tennis

Neural Nets and Decision Tree Handout Problems 2 and 3 due today

4/30

Robot Lab

All three neural network programs due

5/5

Game Playing FOR VIDEO CLICK HERE

Lecture Notes

DUE 5/12 For exam 2 review do problems 1, 4, 5 and 6 on exam2hw.

Read the paper Deep Blue by Murray Campbell, A. Joseph Hoane, and Feng-hsinung Hsu. You can find the paper in the documents section of our Blackboard site.

Use the discussion board in Blackboard to respond to the following question:

Do you think that Deep Blue had an unfair advantage over Kasparov?

Post your position to the discussion board. Use facts from the paper to support your opinion. You may also use information from deep blue's web site.

The official Deep Blue web site: http://www.research.ibm.com/deepblue/home/html/b.html

In addition respond to two of your classmate’s posts. Therefore the minimum number of posts is 3 to receive full credit. Feel free to make additional posts.

DUE 5/19 Extra Credit I - 2 percentage points added to your grade (May substitute for a missing project) Show the calculations and the resulting decision tree given by the ID3 algorithm when given the following training data. The target concept is giveloan

income	credit card	ownshome	married	giveloan
<30	yes	yes	yes	yes
30-80	yes	no	no	no
>80	yes	no	no	yes
<30	no	no	yes	no
>80	no	no	yes	yes
>80	no	no	yes	yes
30-80	no	yes	yes	yes
30-80	yes	no	no	no
30-80	yes	no	no	no
<30	yes	no	yes	no
30-80	no	yes	yes	yes

What rules are defined by the decision tree??

DUE 5/19 Extra credit II - (1 percentage point added to grade May substitute for a missing project)) - Do either A, B OR C. Do not do all three. You will receive credit only for one of them.

Choice A - C4.5 is a decision tree algorithm that implements the methods we discussed. Download and run C4.5 on the play tennis data discussed in class and the restaurant data on page 656 in your text. Print and hand in the results. C4.5 can be found at: http://www.rulequest.com/Personal Note: C4.5 is a UNIX program and you will need a UNIX account to run the software. If you are interested in this project see me so I can get you an account if you don't already have one.

Choice B - The commercial windows version of C4.5, called See5, is available for download at: http://www.rulequest.com/download.html It is limited version of the full software. This means that it will not work with data that has more than 400 records. Notwithstanding, this is sufficient for the play tennis and the restaurant data. Download See5 and print the results from those two datasets

Choice C – Use Weka’s J4.8 algorithm to do evaluate the two data files. It can be found at:

http://www.cs.waikato.ac.nz/ml/weka/

5/7

Neural Network Programs due into digital Drop Box

Class Cancelled

5/12

Discussion Board Posts due.

Exam2hw due

Regular Class - Game Playing

For Video Click Here

5/14

Final Exam