Lecture Slides
- Lecture 1 -- Introduction to AI and agents
- Lecture 2 -- Basic search.
- Lecture 3 -- Informed and local search
- Lecture 4 -- Constraint-satisfaction problems.
- Lecture 5 -- Games and Adversarial Search
- Lecture 6 -- Propositional logic.
- Lecture 7 -- PL inference. First-Order Logic.
- Lecture 8 -- FOL inference.
- Lecture 9 -- Forward and backward chaining. Knowledge Representation. Intro to Natural Language Processing.
- Lecture 10 -- Classical planning.
- Lecture 11 -- Uncertain environments. Bayes nets.
- Lecture 12 -- Approximate inference in Bayes nets. Temporal reasoning under uncertainty.
- Lecture 13 -- HMMs. Decision-making under uncertainty.
- Lecture 14 -- Reinforcement Learning.
- Lecture 15 -- Bayesian learning.
- Lecture 16 -- Neural networks.
- Lecture 18 -- Machine vision.
- Lecture 19 -- Game theory and mechanism design.
- Lecture 20 -- Behavior-based AI and robotics.
Assignments
Review Materials
- Solutions to midterm 2.
- Midterm 2 review problems and solutions
- A good explanation of Bayes theorem and its application.
- Solutions to midterm 1.
- Midterm 1 review and answers
- More review: Search, Games and CSPs - MIT OpenCourseWare, Quiz 1
- More review: Logic - Old Exams on AIMA at Berkeley - Fall '05, Spring '05, questions 3 and 4, Spring '04, questions 4-6.
Other Resources
Readings related to behavior-based AI:- Brooks, R.A. Intelligence Without Representation. Artificial Intelligence 47 (1991), pp. 139-159
- Brooks, R.A. A Robust Layered Control Architecture for a Mobile Robot. IEEE Journal of Robotics and Automation 2:1, March 1986, pp.14-23. Also MIT AI Memo 864, Sept. 1985.
- Sutton and Barto. Reinforcement Learning: An Introduction, 1998. -- A good introductory text on Markov Decision Processes and Reinforcement Leanring.
- The RN AIMA textbook webpage and source code.
The online code repository has code in Java, Python, and Lisp. If you're familiar with C++, I strongly recommend learning Python and using the available code. Python is very easy to learn - you should know most of it after an hour-long tutorial - and has many advantages over C++. Here are some Python resources:
- Dive into Python and links therein
- Guido van Rossum's Python Tutorial
Some general-interest links related to the history and practice of AI:
- Vintage video explaining the control (with STRIPS programmed in Lisp) of the Shakey robot.
- Alan Turing (1950) Computing Machinery and Intelligence. Mind 59, 433-460.
- The Loebner Prize is awared to the system that comes closest to passing a variant of the Turing Test each year.
- An interesting inverse of the Turing Test at the end of this paper by Douglas Hofstadter: A Coffeehouse Conversation on the Turing Test