claimroutineinferred_dep
Coupling tail bounds uniform in history
Statement (paper)
There exist fixed constants and such that, for every finite history map on , under the joint law with an independent uniformly random embedding , the probability that any of has embedded absolute value greater than is less than , and the probability that any of the events or occurs is less than in total.
In plain English
There are fixed constants and such that, for every finite history map on , under the coupled law with an independent uniform random embedding , all the listed normalized coordinates and edge ratios have embedded absolute value at most except with probability under . Likewise, the events that distinct values among or among have embedded difference smaller than have total probability under .
Why it's here
The bounded mean heights from the previous claim do not depend on the history. Combining them with the Markov-type tail bounds of the height lemma, a union bound over the finitely many listed quantities gives large-value and near-coincidence bounds that are uniform in . This gives the next claim a good set of outcomes where the tests pass and the values stay separated.
Classified routine:
Follows from the history-independent bounded mean heights of the previous claim combined with the Markov-type tail bounds from the height lemma, via a union bound over finitely many quantities.
Uses
Used by
Sources
PaperOpen
- Section
- §4.3 Excluding bounded total variation
- Page
- —
- Label
—- Line
sections/03-heights.tex:250
Leannot aligned
Phase 2 aligns this node to a Lean declaration.
Friction
- infoinferred_depoutline: "Coupling tail bounds uniform in history" rests on claim-ht-bounded-1
- infoinferred_depoutline: "Coupling tail bounds uniform in history" rests on ht:height