G.9:8 — Archetypal grounding

Preface node heading:g-9-8-archetypal-grounding:140914

What this page is

This is generated FPF reference text from the specification preface or supporting sections. It helps interpret FPF; it is not FPF Reference product documentation.

Methodology

Use it to understand how the specification wants to be read, then return to a route, pattern, or work packet for active work. Cite generated IDs only when the wording changes the task decision.

Content

Show-A — Multi-tradition parity for decision systems. ParityPlan pins the evidence window and comparator references. ParityReport publishes what the comparison established and the supporting evidence. A commissioned selection adds its declared outcome. Preference-learning comparators, causal decision pipelines, offline-RL evaluation pipelines and robust BO-style selectors remain illustrative families until a G.2 SoTA pack or named current source identifies the family actually compared.

Show-B — QD parity (for example, CMA-MAE arXiv:2205.10752 and DQD arXiv:2106.03894). ParityPlan pins descriptor/distance definitions and archive insertion policy editions. ParityReport identifies the compared archive states, records the lawful comparison result and telemetry deltas, and keeps illumination summaries separate from dominance unless explicitly promoted.

Show-C — Open-ended parity (POET arXiv:1901.01753 as an environment–method co-evolution lineage). ParityPlan pins transfer rule editions and exploration policy refs. ParityReport publishes the comparison and any declared selected-set outcome with transfer-keyed traces (PathSlice). A policy change identifies the dependent uses for G.11 to reconsider.

Show-D — Causal method rung parity. A team compares an observational predictor, an intervention optimizer, and a counterfactual policy strategy under one "best causal method" headline. The team first applies CausalRungParityScreen. If differences in rungs, support components, estimands, endpoints or outcome windows make the proposed comparison untenable, it reports degraded parity or abstention without opening the full record.

When full parity remains plausible, G.9 requires CausalMethodRungParityRecord. For each method, the record states targetCausalUseClaimKind, target CausalityLadderRung, causalEstimandRef, interventional-action basis, the support components consumed, relevant C.28 support result, follow-up window, outcome measure, changed transport endpoints, and estimate-result basis.

When fields differ, the report names the applicable declaredCausalityLadderBridgeOrLossRef and any available transportability or estimation references. It reports degraded parity or abstention where parity cannot be established. The admissible output may be a selected set by comparable rung, not one scalar winner.


Last Updated: 2026-11-22 — upstream FPF commit 7a93d8f8 (github.com/ailev/FPF)