Universal Artificial Intelligence: Sequential Decisions Based on Algorithmic Probability

Author: Marcus Hutter
Links: Book site

Summary

Defines AIXI — a formal model of optimal AGI that combines Solomonoff induction with sequential decision theory. Proves optimality bounds for a theoretical agent that maximizes expected reward.

Relevance to Hyperon

AIXI provides the theoretical upper bound that Hyperon's practical architecture aspires toward. Understanding where and why Hyperon departs from AIXI-optimality helps clarify architectural design decisions.