AFTD Auto-Formalizing Theoretical Domains

spi_two_point_pool_prob

def

The probability with which the low value joins the accepted signal in the optimal scheme: 1 if T <= E[X], q(h-T)/((1-q)(T-l)) if E[X] < T <= h, 0 if T > h.

Statement

noncomputable def spi_two_point_pool_prob (h l q T : ℝ) : ℝ
Defined in
Intrinsic Robustness of Prophet Inequality to Strategic Reward Signaling, NeurIPS 2024 (arXiv:2409.18269), Proposition 3.1 (threshold signaling: pooling probability with T = E[X | accepted]); two-point case
Verified
06 Oct 2026
Axioms
none listed
Used by 4
  • spi_two_point_schemeThe optimal scheme of Proposition 3.1 for a two-point reward embedded in S signals: high value sends signal 0, low value sends signal 0…
  • spi_two_point_scheme_prob_eqReal-arithmetic part of the optimality of the two-point scheme: its two signals give acceptance probability exactly spi_two_point_max_prob.
  • spi_two_point_scheme_specFor S >= 2 the two-point scheme is a valid scheme and a player using it is accepted with probability exactly spi_two_point_max_prob.
  • spi_two_point_pool_prob_memThe pooling probability lies in [0, 1].

Lean source view module on GitHub

/-- The probability with which the low value of a two-point reward joins the pooled (accepted) signal in the optimal scheme of Prop. 3.1: `1` if `T ≤ E[X]`, `q (h - T) / ((1 - q) (T - l))` if `E[X] < T ≤ h`, and `0` if `T > h`. -/
noncomputable def spi_two_point_pool_prob (h l q T : ℝ) : ℝ :=
  if T ≤ q * h + (1 - q) * l then 1 else if T ≤ h then q * (h - T) / ((1 - q) * (T - l)) else 0
See it in the graphDiscuss this resultChallenge it