Universal Sequential Decisions in Unknown Environments

TL;DR

This paper introduces the AIXI model, a theoretical framework that unifies decision-making and prediction in unknown environments, addressing limitations of previous models for active agents and passive prediction.

Contribution

It presents a unified model that combines sequential decision theory and Solomonoff induction, extending their applicability to active agents in unknown environments.

Findings

AIXI provides a theoretical foundation for universal decision-making in unknown environments.

The model overcomes limitations of prior theories by integrating active decision-making with universal prediction.

It offers a comprehensive approach to artificial intelligence grounded in formal theory.

Abstract

We give a brief introduction to the AIXI model, which unifies and overcomes the limitations of sequential decision theory and universal Solomonoff induction. While the former theory is suited for active agents in known environments, the latter is suited for passive prediction of unknown environments.