OpenAI preprint, September 2026 · §5. Homogeneous increments and signed Poisson sampling
§5 (setting): the power-law intensity and signed sums:Homogeneous integration and logarithmic moments accompany convolution powers and compound Poisson measures
Power intensities and signed jump sums.
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
Signed sampling for real spaces involves a measure on positive jump magnitudes and measures of sums of real vector-valued jumps. Let be the real additive space in which the jumps take values, and write for convolution of measures on .
The Claim
Definition 1. For , the power intensity is the Borel measure on determined by (1) for every Borel set , where :(1)✦
The symmetrization of a measure on is the measure in (2), where and denotes the pushforward of by :(2)✦
The convolution powers of a measure on are the measures specified by (3), with the unit point mass at zero: (3)✦
The compound Poisson measure on , with rate and jump measure , is the measure defined by (4): (4)✦
Theorem 1. Let .For any function , where is a real normed vector space, integration against is given by (5): (5)✦For , the tail is given by (6): (6)✦For and an even function with , the scaling identity is (7): (7)✦For , the critical truncated moment is (8): (8)✦
What Is Checked
The measure constructions are stated over arbitrary jump measures and nonnegative rates; the integral identities concern positive exponents and real normed target spaces, with even scalar functions vanishing at zero in the scaling identity.