OpenAI preprint, September 2026 · §3. Common mass, smoothing, and score moments
Lemma3.1:Independent copies of a centered finite random variable with a bounded even moment
Coefficient energy bounds multilinear even moments.
Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its
declaration both in Lean and in words.
The Setting
A finite probability law on a finite outcome set is a mass function satisfying (1). (1)✦The expectation of a real function on is the weighted sum in (2). (2)✦
For an integer , let .The product law on is the probability law of independent samples from , with masses given by (3). (3)✦
For a real coefficient array on the subsets of and a function , the multilinear polynomial is the function in (4). (4)✦The weighted coefficient energy , for , is the counting-measure sum in (5). (5)✦
The squared norm of a real function on the outcome set of a finite probability law , with an integer, is the probability-weighted moment in (6).✦For a real function on the outcome set of a finite probability law , the even-moment quantity at is the real number .(6)✦
The Claim
Theorem 1. Let be an integer, and let be centered under , so that . Let be an integer and a real coefficient array on the subsets of .If , the even moment obeys (7). (7)✦If and , the squared norm obeys (8). (8)✦
In the paper, the result reads: for an even integer ,, and independent centered real random variables with , every integer and real coefficient array on the -subsets of satisfy , with absolute.
What Is Checked
The claims are stated over finite probability laws and their product laws, arbitrary real coefficient arrays indexed by subsets, and centered real functions with the stated even moment bounds.