Algorithmic Complexity Theory

Duane A. Bailey
This lecture concerns algorithmic complexity theory, or Kolmogorov Complexity. The essential feature of this theory is: an object is no more complex than the shortest program that accurately describes it.
Outline of this class (resources are below)
  1. Review of Shannon's notion of entropy. Entropy in a communication channel translates to uncertainty in the values soon to be transmitted. This suggests that there is a current lack of information that will be satisified by the incoming message.
  2. Notion of randomness: a system with high entropy. Hard to construct the exact state because all features appear unrelated.
  3. A bit of programming.
  4. Kolmogorov complexity: the shortest program that describes an object, determines its complexity.
Resources needed for next class: These items require Acrobat Reader.
These items require real player or windows media player
These items are Java applets, and require Java to be installed on your machine.