Course Details

 

Text:

            AI: A Modern Approach, Stuart Russell and Peter Norvig,
            Additional Readings: Marr, Bishop, occasionally others.

 

Prerequisites:

           CS 210 Data Structures.

 

Course Objective:

This course is a project-oriented course in which only the very fundamental material will be covered. You are then expected to select a project to investigate further. The class will also learn from your work on this project, which may be either an application or a theoretical topic.

 

Grades:

Two exams - 15% each
Homework and Labs - 20%. Expect to work hard on these.
Final Project - 50% (Breakup: Proposal: 5%, Presentation 10%, Report 15%, Demo/Oral 20%)

 

Projects
 
Projects from past years may be seen at http://web.cse.iitk.ac.in/~cs365/projects.html
Some projects with a Machine Learning focus may also be relevant: http://web.cse.iitk.ac.in/~cs674/projects.html
Owing to the high project weightage, project groups will be formed by lottery within the first two weeks of class.


COURSE OUTLINE

Topic Week Ref
INTRO: The past and the future: AI is "anything that the computers can't do... yet." Cognitive Science and AI lecture 1 ch.1
AGENTS: Rational Agents --> Learning Agents --> Cognitive Agents. Dynamic vs Static. Learning. Sensor-Action maps. week 2 ch.2
SEARCH and CONSTRAINT PROPAGATION: Game playing week 3-4 ch.3,5,6
STUDENT PRESENTATIONS ON LITERATURE week 4 ch.
LEARNING. Learning Logical Structures: Decision Trees week 5 ch.18,7
PROBABILISTIC Reasoning: Bayesian Networks week 6-7 ch.13,14,
bishop.1
SENSING: Vision - Image Formation, Gradient and Motion cues, Learning Backgrounds, Tracking

 week 8-9 ch.24,
marr.1
ACTION: Robotics - articulated and mobile robots; motion planning, task planning.

 week 10 ch.25
PROBABILISTIC LEARNING: Neural Networks and SVMs week 11-12 ch.