The New AI: General & Sound & Relevant for Physics

TL;DR

This paper discusses recent advances in AI algorithms that are theoretically optimal and practically feasible, with implications for physics and the universe's computational nature.

Contribution

It highlights the development of general, sound, and relevant AI algorithms that extend beyond traditional heuristics and limited systems.

Findings

Progress in prediction, search, and decision algorithms

Relevance of AI to physics and universe modeling

Implications of computer-generated universe hypothesis

Abstract

Most traditional artificial intelligence (AI) systems of the past 50 years are either very limited, or based on heuristics, or both. The new millennium, however, has brought substantial progress in the field of theoretically optimal and practically feasible algorithms for prediction, search, inductive inference based on Occam's razor, problem solving, decision making, and reinforcement learning in environments of a very general type. Since inductive inference is at the heart of all inductive sciences, some of the results are relevant not only for AI and computer science but also for physics, provoking nontraditional predictions based on Zuse's thesis of the computer-generated universe.