Alphabeta Math
CounterexampleConstruction: AI-adaptedVerification: AI-adaptedPipeline-generatedprecheck passjudge pass (gpt-6-sol)audited 2026-09-30
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Different classes can have different recurrence types

Statement refuted

The false assertion is that recurrence or transience must be shared by all states in a Markov chain, including states in different communicating classes.

Facts & Assumptions

Given: The two-state space E={0,1} and transition matrix p(0,0)=1,p(0,1)=0,p(1,0)=12,p(1,1)=12.

[F1]

States communicate exactly when each is accessible from the other, and x→y means p(n)(x,y)>0 for some n≥0. (Accessibility, communication, and irreducibility)

[F2]

The zero-step matrix is p(0)(x,y)=1{x=y}. (Transition matrices and n-step probabilities)

[F3]

The matrix powers satisfy p(m+n)(x,y)=∑z∈Ep(m)(x,z)p(n)(z,y) for m,n≥0. (Matrix Chapman–Kolmogorov equations)

[F4]

The positive return time is Tx+=inf⁡{n≥1:Xn=x}. (Hitting, return, and visit times)

[F5]

State x is recurrent if Px(Tx+<∞)=1 and transient if this probability is less than one. (Recurrent and transient states)

Counterexample

1.1given

The entries of p are nonnegative and its two row sums are 1+0=1 and 12+12=1, so the displayed table is a stochastic matrix.

1.2F1F2F3given

The row from 0 is concentrated at 0. By induction using [F3], p(n)(0,0)=1 and p(n)(0,1)=0 for every n≥0. Meanwhile p(1)(1,0)=12>0, so 1→0 but 0↛1. By [F1] and the zero-step identity [F2], the communication relation on these two states is equality; hence its two communicating classes are {0} and {1}.

1.3F4F5given

From state 0, the chain stays at 0 at every step because p(0,0)=1. Thus T0+=1 almost surely and P0(T0+<∞)=1; state 0 is recurrent by [F5].

1.4F4F5given

From state 1, T1+=1 on the first-step transition to 1, which has probability 12. On the other first-step transition, X1=0 and the chain then stays at 0, so T1+=∞. Hence P1(T1+<∞)=12<1, and state 1 is transient by [F4, F5, given]. This verifies the claimed difference in recurrence type across the two distinct classes.

2.1F1F2F4F5step 1.2step 1.3step 1.4given∎

The witness has two states, so the empty-space and one-state cases cannot arise here. The zero transition p(0,1)=0 is essential to the class separation, and the absorbing row at 0 supplies the no-return branch from 1. The return time starts at n=1, so the initial visit at time zero is not counted. The computation uses only the explicit finite transition table and no choice function; it proves a one-way counterexample, not an iff statement.

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

12 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources