On the Foundations of Universal Sequence Prediction

Solomonoff completed the Bayesian framework by providing a rigorous, unique, formal, and universal choice for the model class and the prior. We discuss in breadth how and in which sense universal (non-i.i.d.) sequence prediction solves various (philosophical) problems of traditional Bayesian sequence prediction. We show that Solomonoff’s model possesses many desirable properties: Fast convergence and strong bounds, and in contrast to most classical continuous prior densities has no zero p(oste)rior problem, i.e. can confirm universal hypotheses, is reparametrization and regrouping invariant, and avoids the old-evidence and updating problem. It even performs well (actually better) in non-computable environments.

• Publication year

  2006

• Venue

  Theory and Applications of Models of Computation

• Publication date

  2006-05-03

• Fields of study

  Mathematics, Philosophy, Computer Science

• Identifiers
  DOI 10.1007/11750321_39arXiv cs/0605009
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Sequential Decisions based on Algorithmic Probability

  2008cited by this paper
• Wallace CS: Statistical and inductive inference by minimum message length

  2006cited by this paper
• Statistical and Inductive Inference by Minimum Message Length (Information Science and Statistics)

  2005cited by this paper
• On the Convergence Speed of MDL Predictions for Bernoulli Sequences

  2004cited by this paper
• Universal Convergence of Semimeasures on Individual Random Sequences

  2004cited by this paper
• Convergence and Loss Bounds for Bayesian Sequence Prediction

  2003cited by this paper
• Clustering by compression

  2003cited by this paper
• Optimality of universal Bayesian prediction for general loss and alphabet

  2003cited by this paper
• Optimal Ordered Problem Solver

  2002cited by this paper
• Hierarchies of Generalized Kolmogorov Complexities and Nonenumerable Universal Measures Computable in the Limit

  2002cited by this paper
• Book Review: An introduction to Kolmogorov Complexity and its Applications Second Edition, 1997 by Ming Li and Paul Vitanyi (Springer (Graduate Text Series))

  1997cited by this paper
• The Selection of Prior Distributions by Formal Rules

  1996cited by this paper
• Inferences from Multinomial Data: Learning About a Bag of Marbles

  1996cited by this paper
• Bayes or Bust?-A Critical Examination of Bayesian Confirmation Theory.

  1994cited by this paper
• An Introduction to Kolmogorov Complexity and Its Applications

  1993influential reference
• Information-theoretic asymptotics of Bayes methods

  1990cited by this paper
• Complexity-based induction systems: Comparisons and convergence theorems

  1978cited by this paper
• THE COMPLEXITY OF FINITE OBJECTS AND THE DEVELOPMENT OF THE CONCEPTS OF INFORMATION AND RANDOMNESS BY MEANS OF THE THEORY OF ALGORITHMS

  1970cited by this paper
• A Formal Theory of Inductive Inference. Part II

  1964influential reference

• On Reward Binarisation and Bayesian Agents

  2022cites this paper
• Illuminating the Dark or how to recover what should not be seen in FE-based classifiers

  2020cites this paper
• Conditions Under Which One Measure Is a Predictor for Another

  2020cites this paper
• Universal Time-Series Forecasting with Mixture Predictors

  2020cites this paper
• Illuminating the Dark or how to recover what should not be seen

  2018cites this paper
• Improved expert selection model for forex trading

  2018cites this paper
• Optimal Capacity Provisioning for Online Job Allocation With Hard Allocation Ratio Requirement

  2016cites this paper
• Online job allocation with hard allocation ratio requirement

  2016cites this paper
• Online Prediction for Forex with an Optimized Experts Selection Model

  2016cites this paper
• Short Biography

  2016cites this paper
• Learning the Experts for Online Sequence Prediction

  2012cites this paper
• LEARNABILITY IN PROBLEMS OF SEQUENTIAL INFERENCE

  2011cites this paper
• Open Problems in Universal Induction & Intelligence

  2009cites this paper
• Predicting non-stationary processes

  2008cites this paper
• Limits of Learning about a Categorical Latent Variable under Prior Near-Ignorance

  2008cites this paper
• Foreword re C. S. Wallace

  2008cites this paper
• Limits of Learning about a Categorical Latent Variable under Prior Near-Ignorance

  2007cites this paper
• On Sequence Prediction for Arbitrary Measures

  2006cites this paper
• Is There an Elegant Universal Theory of Prediction?

  2006cites this paper