Text: Artificial Intelligence: A Modern Approach, Second Edition, (Prentice-Hall, ISBN 0-13-790395-2)
Authors: Stuart Russell and Peter Norvig
This is a combined senior-level and graduate course in artificial intelligence (AI), a computational study of intelligent behavior. The focus will be on intelligent agents, which could be software agents or robots. How agents decide what to do, and how they learn from experience in the world, will be covered.
COSC 3020.
I. AGENTS:
Introduction to AI (Chapter 1)
Agents (Chapter 2)
Robots (25.1-25.2)
Multiple Agents
II. MACHINE LEARNING FOR AGENTS:
Reinforcement learning (Read 21.3, but since we are jumping ahead, don't worry if you don't understand it all. The lecture should clarify the material.)
III. DISCUSSION OF TERM PROJECT TOPICS
IV. AGENT EVALUATION, PART I:
Background: Probability (13.1-13.5)
Experimental methodology
V. SEARCH:
Search (3.1-3.3)
Uninformed search (3.4-3.7)
Informed (heuristic) search (4.1-4.2)
Hill climbing; Stochastic search (Simulated annealing, genetic algorithms) (4.3-4.6)
VI. APPLICATIONS OF SEARCH:
Intro to planning.
Intro to games.
VII. STOCHASTIC MODELS:
Background: Conditional probability (13.1-13.8)
Bayesian (belief) networks (Chapter 14)
VIII. NEURAL MODELS:
Neural networks and perceptrons (20.5, but we will not cover neural network learning)
IX. AGENT EVALUATION, PART II:
Statistical significance tests
X. IF THERE IS TIME REMAINING, WE WILL DISCUSS INTERESTING TOPICS ON AI
For students who are interested in studying more about robotic and software agents, Dr. Spears also teaches another course on this topic: Modern Robots and Softbots, course syllabus.
For students who are interested in studying more about machine learning, Dr. Spears teaches a course on this topic: Machine Learning course syllabus.
One term project is required, to be done as part of a team. You need to start as early as possible and work steadily on it all semester in order to finish in time.
Term project proposals: due date TBD.
Final term project presentations and demos: due date TBD. A final writeup of the project is required. A sample writeup is a handout.
There will be multiple written and programming homework assignments, to be done individually. Late homework will lose 20% credit each class period it is late.
There will be a midterm and a final exam.
The various required work in this class will be counted towards your final grade as follows:
Term project - 35%
Written homework - 20%
Midterm exam - 20%
Final exam - 25%
Grading will be on a curve. Work is due at the beginning of class, and late work is accepted for a few days, or until a solution is distributed, at a substantial reduction in credit (see above). Returned work should be kept for verification of records.
The professor reserves the right to alter the grading scheme or to take extenuating circumstances into account when assigning grades. Discussion of the course material among students is encouraged, although students are expected to write up their own assignments and programs. Copying code, homework solutions, or exam material from any source at all (unless explicitly permitted by the instructor) will most likely result in failing the course. Academic dishonesty will be treated in accordance with the strictest university standards. Students are urged to read University regulation 803 , section 3 defines academic dishonesty.