Result · PROXY-TRUE-RETAINED-FI
Proxy-built labels retain the true score's information, not the proxy's
Derived from the literaturepublication: derived from published
Meaning
The true score, conditioned on proxy-derived labels, determines achieved Fisher information; proxy moments measure a different quantity.
Scientific context
Every theorem in this area is a statement about a table of numbers; it is silent about where those numbers came from. When the score is estimated — from a classifier, a learned ratio, a simulator — the information actually retained is governed by the true score conditioned on the estimated rule, not by the estimated score conditioned on itself. The two coincide only when the proxy is exact in the relevant sense; this is why estimated-score provenance never claims exact Fisher semantics.
Statement
If q is built from estimated score s_hat, the actual retained information is Var(E[s|q(s_hat)]), not Var(E[s_hat|q(s_hat)]) unless the proxy is exact in the relevant sense.
Relationships
Rests on
Enables
- Reported versus true retention of a frozen label map under a score error: an exact two-term expansion in the Fisher-whitened score errorSCORE-ERROR-RETENTION-BUDGET
- Applying a frozen rule to the proxy instead of the truth: mislabel mass under a margin condition and an exact Loewner sandwich on the retained informationSCORE-ERROR-RULE-TRANSFER
- Classifier calibration error to Fisher-information lossOPEN-CLASSIFIER-CALIBRATION-FI
- Finite-sample uncertainty for retention estimatesOPEN-RETENTION-UNCERTAINTY
- Perturbation theory for estimated scoresOPEN-SCORE-PERTURBATION
- Frozen-rule scalar retention: plug-in CLT with a consistent influence-function varianceRETENTION-PLUGIN-CLT-FROZEN-SCALAR
- Frozen-rule vector retention: plug-in CLT for the geometric-mean D-retention with a consistent influence-function varianceRETENTION-PLUGIN-CLT-FROZEN-VECTOR
Raises
Open next door
- Classifier calibration error to Fisher-information lossOPEN-CLASSIFIER-CALIBRATION-FI
- Perturbation theory for estimated scoresOPEN-SCORE-PERTURBATION
- Direction-wise guarantees from determinant efficiencyOPEN-D-DIRECTIONAL-BOUND
- Sharp D-efficiency bounds versus KOPEN-D-EFFICIENCY-VS-K
- Estimate representation versus quantization lossOPEN-REPRESENTATION-LOSS-ESTIMATION
- Finite-sample uncertainty for retention estimatesOPEN-RETENTION-UNCERTAINTY
In the library
Carried by RatioProvenance, ScoreProvenance. Reference (opens in a new tab).
Local graph
Proof
O4. True retained FI under an estimated score
If the quantizer uses , the actual retained Fisher information is
not unless in the relevant sense.