Techniques in Artificial Intelligence (SMA 5504)
Graduate-level AI: first-order logic, planning, Bayesian networks and supervised learning, with notes, exams, exercises and Java programming.
This graduate-level introduction to artificial intelligence concentrates on how machines represent knowledge and reason under certainty and uncertainty. It covers representation and inference in first-order logic, deterministic and decision-theoretic planning, basic supervised learning methods, and inference and learning in Bayesian networks. It was also taught as part of an alliance with Singapore. Assignments are programmed in Java, and past exams support review.
What You Will Achieve
- Represent and reason with first-order logic
- Apply deterministic and decision-theoretic planning techniques
- Perform inference and learning in Bayesian networks
- Use basic supervised learning methods
Program Overview
Search and Logic
Reasoning with certain knowledge.
- Search: A* and stochastic methods
- Propositional satisfiability
- First-order logic and resolution
Planning
Choosing actions to reach goals.
- Situation calculus and POP
- GraphPlan and SATPlan
- Markov decision processes
- Probabilistic planning
Probability and Learning
Reasoning and learning under uncertainty.
- Bayesian networks
- Decision theory
- Reinforcement learning
- Supervised learning
Course Completion
Job Outlook
This grounding is useful for further study and research in artificial intelligence and machine learning.
Good for: Computer science students who want a rigorous grounding in classical logic, planning and probabilistic methods.
Keep in mind: This 2002 class uses the first edition of its textbook and stops at basic supervised learning, so it says nothing about later methods. Past exams and many exercise solutions are available.
Bottom line: A sound course on classical methods in the field, with plenty of practice material.
Frequently Asked Questions
What textbook does it use?
Russell and Norvig's Artificial Intelligence: A Modern Approach, first edition, with some extra readings, including one chapter from the second edition, and survey papers.
What programming is involved?
Course assignments require Java. The assignments section also links excerpts of code, documentation and downloads used to complete the work.
What exam practice is available?
Final exams from Fall 2001 and Spring 2001, a sample final and a practice midterm are published as study aids, along with exercises for many lectures.



There are no reviews yet.