Pith. sign in

REVIEW

Upper tails for arithmetic progressions revisited

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2409.08383 v1 pith:DQ3T5RO6 submitted 2024-09-12 math.PR math.CO

Upper tails for arithmetic progressions revisited

classification math.PR math.CO
keywords arithmeticprobabilityprogressionsrandomupper-tailappearanceargumentsasymptotically
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Let $X$ be the number of $k$-term arithmetic progressions contained in the $p$-biased random subset of the first $N$ positive integers. We give asymptotically sharp estimates on the logarithmic upper-tail probability $\log \Pr(X \ge E[X] + t)$ for all $\Omega(N^{-2/k}) \le p \ll 1$ and all $t \gg \sqrt{Var(X)}$, excluding only a few boundary cases. In particular, we show that the space of parameters $(p,t)$ is partitioned into three phenomenologically distinct regions, where the upper-tail probabilities either resemble those of Gaussian or Poisson random variables, or are naturally described by the probability of appearance of a small set that contains nearly all of the excess $t$ progressions. We employ a variety of tools from probability theory, including classical tilting arguments and martingale concentration inequalities. However, the main technical innovation is a combinatorial result that establishes a stronger version of `entropic stability' for sets with rich arithmetic structure.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.