Predictions and algorithmic statistics for infinite sequence

TL;DR

This paper introduces a new prediction method for infinite binary sequences that overcomes limitations of Solomonoff's approach, providing similar theoretical guarantees while avoiding some of its negative aspects.

Contribution

The authors propose a novel prediction technique based on the best distribution for each string, with proven theoretical properties similar to Solomonoff's but without its negative issues.

Findings

Proves a Solomonoff-like theorem for the new prediction method.

Shows the new method avoids the divergence issues of Solomonoff's approach.

Provides theoretical guarantees for the new prediction scheme.

Abstract

Consider the following prediction problem. Assume that there is a block box that produces bits according to some unknown computable distribution on the binary tree. We know first $n$ bits $x_1 x_2 \ldots x_n$. We want to know the probability of the event that that the next bit is equal to $1$. Solomonoff suggested to use universal semimeasure $m$ for solving this task. He proved that for every computable distribution $P$ and for every $b \in \{0,1\}$ the following holds: $$\sum_{n=1}^{\infty}\sum_{x: l(x)=n} P(x) (P(b | x) - m(b | x))^2 < \infty\ .$$ However, Solomonoff's method has a negative aspect: Hutter and Muchnik proved that there are an universal semimeasure $m$, computable distribution $P$ and a random (in Martin-L{\"o}f sense) sequence $x_1 x_2\ldots$ such that $\lim_{n \to \infty} P(x_{n+1} | x_1\ldots x_n) - m(x_{n+1} | x_1\ldots x_n) \nrightarrow 0$. We suggest a new way for prediction. For every finite string $x$ we predict the new bit according to the best (in some sence) distribution for $x$. We prove the similar result as Solomonoff theorem for our way of prediction. Also we show that our method of prediction has no that negative aspect as Solomonoff's method.