Universal prediction, i.e. Bayesian framework + Solomonoff prior + model class of all computable sequences, solves/avoids foundational problems of inductive inference, but has to be compromised in practice.

Why is our world compressible? How is induction possible? What is the relation between physical and algorithmic entropy?

Videos and slides on algorithmic complexity, universal induction and learning theory.

Suggests a method to obtain a stable approximation of the Kolmogorov complexity for sequences generated by algorithms which are shorter than typical compiler lengths.

Applies basic concepts of Kolmogorov compexity theory to the set of possible universes. Ideas on life, true randomness, generalization, and learning in a given universe.

The idea behind Kolmogorov complexity: a good measure of the complexity of an object is the length of the shortest computer program which will construct that object. From this basic idea a variety of insights and powerful techniques have been developed.

An infinite sequence S is random if and only if no prefix can be produced by a program much shorter than the prefix.