Universal artificial intelligence. Sequential decisions based on algorithmic probability.

DOI10.1007/b138233zbMath1099.68082OpenAlexW2013391942WikidataQ58012464 ScholiaQ58012464MaRDI QIDQ1762575

Marcus Hutter

Publication date: 11 February 2005

Published in: Texts in Theoretical Computer Science. An EATCS Series (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/b138233

zbMATH Keywords

algorithmic information theorysupervised learningfunction minimizationstrategic gamessequence predictionsequential decision theoryunknown environmentreinforcement learning agent

Mathematics Subject Classification ID

Introductory exposition (textbooks, tutorial papers, etc.) pertaining to computer science (68-01) Learning and adaptive systems in artificial intelligence (68T05) Algorithmic information theory (Kolmogorov complexity, etc.) (68Q30) General topics in artificial intelligence (68T01)

Related Items

Reflective Oracles: A Foundation for Game Theory in Artificial Intelligence ⋮ Tractability of batch to sequential conversion ⋮ On semimeasures predicting Martin-Löf random sequences ⋮ Extreme state aggregation beyond Markov decision processes ⋮ On generalized computable universal priors and their convergence ⋮ Algorithmic complexity bounds on future prediction errors ⋮ Open problems in universal induction \& intelligence ⋮ Obituary: Ray Solomonoff, founding father of algorithmic information theory ⋮ A complete theory of everything (will be subjective) ⋮ Universal knowledge-seeking agents ⋮ General time consistent discounting ⋮ DECISION TREES DO NOT GENERALIZE TO NEW VARIATIONS ⋮ A philosophical treatise of universal induction ⋮ A generalized characterization of algorithmic probability ⋮ Predicting non-stationary processes ⋮ Approval-directed agency and the decision theory of Newcomb-like problems ⋮ Using ideas of Kolmogorov complexity for studying biological texts ⋮ On the computability of Solomonoff induction and AIXI ⋮ On Universal Transfer Learning ⋮ Some Sufficient Conditions on an Arbitrary Class of Stochastic Processes for the Existence of a Predictor ⋮ The Quest for Uncertainty ⋮ A Criterion for Optimal Predictive Model Selection ⋮ Reward is enough ⋮ On the possibility of learning in reactive environments with arbitrary dependence ⋮ The probability of ``undefined (non-converging) output in generating the universal probability distribution ⋮ Time-universal data compression ⋮ Application of Kolmogorov complexity and universal codes to identity testing and nonparametric testing of serial independence for time series ⋮ The Neuronal Replicator Hypothesis ⋮ A theory of incremental compression ⋮ On the Computability of Solomonoff Induction and Knowledge-Seeking ⋮ Stationary algorithmic probability ⋮ Sequential predictions based on algorithmic complexity