Consensus probability (Theorem 2.1) #
Eventual consensus probability is the supremum of increasing finite-time probabilities. The proof bounds the discrepancy from invariant white mass by the probability of nonconsensus, then passes to the limit.
Probability that all vertices have color c at time n.
Equations
- Voter.colorProbability H c n s = (Voter.transition H).iterate n (Voter.allColor c) s
Instances For
Eventual probability, defined from the finite-time consensus probabilities.
Equations
- Voter.eventualColor H c s = ⨆ (n : ℕ), Voter.colorProbability H c n s
Instances For
A constant configuration retains its color at every finite time.
A constant configuration reaches its own consensus color with probability one.
Stationary white mass.
Equations
- Voter.whiteMass p = Voter.mass p fun (b : Bool) => if b = true then 1 else 0
Instances For
Finite-time consensus on a color is an event.
The finite-time discrepancy is at most nonconsensus probability (Lemma 2.2).
Finite-time all-white probability converges to initial stationary white mass.
Hassin–Peleg Theorem 2.1. Eventual all-white consensus probability is exactly the stationary weight of the initially white vertices.