Reproduce “Solomonoff induction of selected bits and online classification” 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
Bruno Bauwens, German Kuznetsov
The authors analyze Solomonoff prediction for binary classification, where inputs may be difficult to model but labels follow a simple rule. A binary classifier with a k-bit program and fixed computable runtime bound has, for any input sequence and each fixed γ>1, a sum of γ-th powers of wrong-label probabilities of O(k^{1+γ} log^{2γ} k). This theoretical guarantee is not a finite bound on absolute error.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.