Algorithmic Information Theory
- Yuki
- Jun 27
- 1 min read
A theory that considers "how short a program can be used to represent a given piece of information."
For example, let's compare the following two:
A 111111111111111111111111111111
this is,
"Repeat 1 30 times"
It can be explained briefly as follows: In other words, the amount of information is small.
B 101001110100101101011000111010
The rules are not very clear, so we have no choice but to write it almost exactly as it is. In other words, there is a lot of information.
The more rules something follows, the shorter its description can be. → Kolmogorov complexity
The less structured something is, the harder it is to compress.
This theory doesn't focus on the form itself, but rather on the "shortness of the hidden rules" that create that form.
Architecture where extremely complex spaces unfold from a small number of generation rules.
Is it a space that can be generated with a short algorithm?
Or is it a space with a complexity that is almost impossible to describe?
Comments