Statistical and inductive inference by minimum message length.
estimationinference from stochastic processesinformation-theoretic topicsnon-Markovian processespredictionrandom fieldsresearch exposition
Research exposition (monographs, survey articles) pertaining to statistics (62-02) Foundations and philosophical topics in statistics (62A01) Statistical aspects of information-theoretic topics (62B10) Non-Markovian processes: estimation (62M09) Inference from stochastic processes and prediction (62M20) Random fields; image analysis (62M40) Artificial intelligence (68T99)
Statistical and inductive inference is the foundation on which much of statistical practice is built. This very significant monograph covers the topics of the Minimum Message Length (MML) principle, a new approach to induction, hypothesis testing, model selection, and statistical inference. MML appears also to provide both a normative and a descriptive basis for inductive reasoning and scientific induction. The first chapter in this book contains an introduction to inductive inference and a new approach based on MML. It is intended to clarify the distinctions between conventional statistical inference and the method developed in this book. The second chapter introduces elementary results of Information Theory and Turing Machine Theory, which are needed to define the notion of ``length of explanation and to sharpen the concepts of ``pattern and ``randomness. The third, fourth and fifth chapters formally develop the new approach to statistical inference. Chapter 3 introduces the Strict Minimum Message Length (SMML) estimators which exactly minimize the expected length of explanations, and derivation of certain relations. Chapters 4 and 5 give approximations to the treatment, leading to useable results. Some inference problems are used in these chapters to illustrate the nature and limitations of exact and approximate treatments. In Chapter 6 techniques are introduced needed for some rather more difficult problems. Chapter 7 gives examples of the use of MML in structural models, such as inference of a regular grammar, classification trees and nets, a binary sequence segmentation problem, learning causal nets. Chapter 8 is speculative, presenting an argument that, while deductive (probabilistic) logic is properly applied in predicting the future state of a system whose present state is partly known, useful assertions about the past state of the system require inductive reasoning for which MML appears well suited. Chapter 9 considers whether scientific enquire can be seen as conforming to the MML principle, at least over the long term. Chapter 10 describes Solomonoff's and Rissanen's Normalized Maximum Likelihood approaches, both of which embody the same ``brief encoding notion as Minimum Message Length but apply it to different ends. This valuable book covers the topics at a level suitable for professionals and graduate students in Statistics, Computer Science, Data Mining, Machine Learning, Estimation and Model-selection, Econometrics, etc.
- Minimum message length shrinkage estimation
- Open problems in universal induction \& intelligence
- A complete theory of everything (will be subjective)
- Automatic identification of rock fracture sets using finite mixture models
- Minimum message length inference of the Poisson and geometric models using heavy-tailed prior distributions
- Summarizing categorical data by clustering attributes
- Simplicity and likelihood: an axiomatic approach
- An unsupervised machine-learning checkpoint-restart algorithm using Gaussian mixtures for particle-in-cell simulations
- Pseudo-likelihood, explanatory power, and Bayes's theorem [Comment on: ``A likelihood paradigm for clinical trials]
- Beyond hybrid generative discriminative learning: spherical data classification
- On universal prediction and Bayesian confirmation
- Suboptimal behavior of Bayes and MDL in classification under misspecification
- On multivariate binary data clustering and feature weighting
- High-dimensional count data clustering based on an exponential approximation to the multinomial beta-Liouville distribution
- Minimum message length analysis of multiple short time series
- Book review of: D. L. Dowe (ed.), Algorithmic probability and friends. Bayesian prediction and artificial intelligence
- Subset selection in linear regression using sequentially normalized least squares: asymptotic theory
- Minimum message length analysis of the Behrens-Fisher problem
- MLE’s Bias Pathology, Model Updated MLE, and Wallace’s Minimum Message Length Method
- The role of Bayesian philosophy within Bayesian model selection
- scientific article; zbMATH DE number 4066122 (Why is no real title available?)
- A philosophical treatise of universal induction
- Krimp: mining itemsets that compress
- Adaptive cluster expansion for the inverse Ising problem: convergence, algorithm and tests
- scientific article; zbMATH DE number 740676 (Why is no real title available?)
- scientific article; zbMATH DE number 1122448 (Why is no real title available?)
- Argument interpretation using minimum message length
- Minimum message length grouping of ordered data
- The Complexity of Strict Minimum Message Length Inference
- Change-point estimation using new minimum message length approximations
- Minimum Message Length and Kolmogorov Complexity
- Refinements of MDL and MML Coding
- Rejoinder
- Rejoinder
- Confidence intervals, significance values, maximum likelihood estimates, etc. sharpened into Occam's razors
- The resampling of entropies with the application of biodiversity
- scientific article; zbMATH DE number 6751371 (Why is no real title available?)
- On the selection of predictors by using greedy algorithms and information theoretic criteria
- From Gödel's incompleteness theorem to the completeness of bot beliefs (extended abstract)
- Bridging algorithmic information theory and machine learning: a new approach to kernel learning
- Testing randomness by Matching Pennies
- An axiomatisation of error intolerant estimation
- Recent advances in principal component analysis for directional data
- Turing tests with Turing machines
- Clustgrams: an extension to histogram densities based on the minimum description length principle
- Solomonoff prediction and Occam's razor
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Discussion of: ``Statistical exploration of the manifold hypothesis
- Reply to the discussion of: ``Statistical exploration of the manifold hypothesis
- Model selection, estimation and forecasting in VAR models with short-run and long-run restrictions
- Minimum message length estimation of mixtures of multivariate Gaussian and von Mises-Fisher distributions
- Minimum message length encoding and the comparison of macromolecules
- Minimum message length estimation using EM methods: a case study
- On the entropy production of time series with unidirectional linearity
This page was built for publication: Statistical and inductive inference by minimum message length.
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5461242)
