Recent Results in Universal and Non-Universal Induction

2006/01/01 by Jan Poland, Poland, Jan
Computer Science · #Algorithms and Data Compression #Bayesian learning #Computability, Logic, AI Algorithms #MDL #Machine Learning and Algorithms #Solomonoff induction #prediction with expert advice #stochastic model selection #universal learning

paper · doi:10.4230/dagsemproc.06051.12

openalex publication_date 2006/01/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present and relate recent results in prediction based on countable classes of either probability (semi-)distributions or base predictors. Learning by Bayes, MDL, and stochastic model selection will be considered as instances of the first category. In particular, we will show how analog assertions to Solomonoff's universal induction result can be obtained for MDL and stochastic model selection. The second category is based on prediction with expert advice. We will present a recent construction to define a universal learner in this framework.

Citations

• The Nonstochastic Multiarmed Bandit Problem

Related

• Solomonoff induction
• On Universal Features for High-Dimensional Learning and Inference
• Is there an Elegant Universal Theory of Prediction?
• Learning Whenever Learning is Possible: Universal Learning under General Stochastic Processes
• Generation through the lens of learning theory