pith. sign in
module module high

IndisputableMonolith.Decision.StPetersburg

show as:
view Lean formalization →

This module sets up the St. Petersburg paradox in Recognition Science decision theory by defining payout 2^n on the nth heads outcome together with probabilities and partial utility sums. It would be cited by researchers examining infinite-expectation paradoxes under RS cost functions. The module contains only definitions and elementary properties of linear and logarithmic partial utilities with no theorem proofs.

claimPayout on outcome $n$ is $p(n)=2^n$ for $ngeq0$. Probability is $q(n)=2^{-n}$. Linear partial utility up to $N$ is the sum of $p(k)q(k)$ terms; logarithmic partial utility is the corresponding sum of $log(p(k))q(k)$ terms. The $n=0$ term contributes zero to both sums.

background

The module imports Constants, whose fundamental object is the RS time quantum $tau_0=1$ tick, and the Cost module that supplies the J-cost function obeying the Recognition Composition Law. It introduces the classic St. Petersburg game: a fair coin is flipped until the first heads appears on flip $n$, at which point the payout is $2^n$. Sibling definitions include prob_zero, prob_pos, linearUtilityPartial, logUtilityPartial, and the auxiliary linear_term and logTerm objects that isolate the divergent and convergent behaviors.

proof idea

this is a definition module, no proofs

why it matters in Recognition Science

The module supplies the concrete decision-theoretic example needed to test how RS-native utilities (linear versus logarithmic) behave on an infinite-expectation problem. It feeds parent results in the broader Decision domain that connect payout and probability structures to the J-uniqueness and phi fixed-point steps of the forcing chain, although the current dependency graph lists no direct used_by edges.

scope and limits

depends on (2)

Lean names referenced from this declaration's body.

declarations in this module (24)