Reproduce “Identification of Probabilities” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Paul M.B. Vitanyi, Nick Chater

The authors show that computable probability models can be identified from examples when candidates belong to effectively listed classes. They show learners can eventually settle on distributions from independent draws and computable ergodic Markov chains almost surely, and on a measure for typical dependent sequences. The stabilization point is unknown, and the dependent result need not identify a unique source.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.