claimroutineinferred_dep
Pseudorandom vertices have local payoff bound
Pseudorandom vertices have payoff ≤ plus small errors
local bound for pseudorandom vertices
Statement (paper)
For every left vertex that is pseudorandom at threshold ,
In plain English
For every left vertex that is pseudorandom at threshold , its payoff under the ordering is at most .
Why it's here
This bounds the contribution of pseudorandom left vertices by a uniform local constant. The block functions are simplex-valued, have mean and low influence, so the rounding lemma gives the term and the collision lemma controls pairs of ordering coordinates, giving the term. The next claim then shows that if the total payoff exceeds this bound, nonpseudorandom vertices must account for the excess, which decodes to a good game labeling.
Classified routine:
Direct combination of the earlier rounding lemma (giving the opt_q(J)+r_s(kqt) term) and the collision lemma applied to pairs of ordering coordinates, using the simplex-valued, low-influence block functions from the previous claim; this is the standard soundness step in Raghavendra/Guruswami et al.-style ordering-CSP reductions from Unique Games.
Uses
Used by
Sources
PaperOpen
- Section
- §A.2 The finite ordering transfer
- Page
- —
- Label
—- Line
appendices/finite-consequences.tex:737
Leannot aligned
Phase 2 aligns this node to a Lean declaration.
Friction
- infoinferred_depoutline: "Pseudorandom vertices have local payoff bound" rests on claim-prop-finite-ordering-transfer-2