An empirical study of the complexity and randomness of prediction error sequences

Joel Ratsaby

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

We investigate a population of binary mistake sequences that result from learning with parametric models of different order. We obtain estimates of their error, algorithmic complexity and divergence from a purely random Bernoulli sequence. We study the relationship of these variables to the learner's information density parameter which is defined as the ratio between the lengths of the compressed to uncompressed files that contain the learner's decision rule. The results indicate that good learners have a low information density ρ while bad learners have a high ρ Bad learners generate mistake sequences that are atypically complex or diverge stochastically from a purely random Bernoulli sequence. Good learners generate typically complex sequences with low divergence from Bernoulli sequences and they include mistake sequences generated by the Bayes optimal predictor. Based on the static algorithmic interference model of [18] the learner here acts as a static structure which " scatters" the bits of an input sequence (to be predicted) in proportion to its information density ρ thereby deforming its randomness characteristics.

Original languageEnglish
Pages (from-to)2832-2844
Number of pages13
JournalCommunications in Nonlinear Science and Numerical Simulation
Volume16
Issue number7
DOIs
StatePublished - 1 Jul 2011
Externally publishedYes

Keywords

  • Algorithmic complexity
  • Binary sequences
  • Chaotic scattering
  • Description complexity
  • Information theory
  • Prediction
  • Statistical learning

ASJC Scopus subject areas

  • Numerical Analysis
  • Modeling and Simulation
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'An empirical study of the complexity and randomness of prediction error sequences'. Together they form a unique fingerprint.

Cite this