nuisance concept
| Definition | The directions of the data a consumer cannot distinguish, the kernel of its read operator. Chapter 1. |
|---|---|
| Example | For the consumer x1 + x2, the direction (1, −1) is nuisance: moving along it changes nothing. |
| Book | Data Mining as Observation, draft 0.2, commit f3914f0; entry id nuisance, kind concept. |
| Status | measures [predicted]. Corrections: none recorded. |
| Defining equation | none |
| Assumptions and scope |
|
| Prior art | none recorded |
| Evidence | geometric-observation/claims/LEDGER.md:62, lean/DataMiningAsObservation/ReadOperator.lean |
| Reviewed | not yet reviewed; generated 2026-09-10 from records at the commits on the provenance page. |
Equation
none
Conditions
- The nuisance is the kernel of the workload-averaged read operator, the set of directions unread at almost every row of that workload, since the kernel of the average is the intersection of the local kernels.
- For a nonlinear consumer a null direction can turn with the row, so the nuisance of the average is not a global equivalence of inputs, and cosine similarity’s radial direction is the case.
- A transform is safe when the reader it produces still resolves the read subspace, and discretization is safe only when the values it merges differ in nuisance directions.
Conditions are curated in entries.toml rather than read
from a record.
Ledger
- measures. GO-1
[predicted]. The consumer’s invariant/nuisance split is identifiable ex ante from the consumer functional.geometric-observation/claims/LEDGER.md:62.
First stated
Volume 14, chapter 5,
geometric-observation/chapters/ch05_the_read_metric_and_the_quotient.md:7-48,
DOI 10.5281/zenodo.21776291, where the nuisance is the kernel of the
read operator.
Measurements
none
Failures and corrections
none
Invariance envelope
Survived.
- OD:observer-family, change of observer within a declared family:
read operators of declared spectrum, coordinate subsets, coarsenings,
and learned consumers’ recovered read operators. Claim: An observer with
a kernel reads a smaller exponent than the classical one when an unread
direction grows faster (World K, mu = 2), the same when it grows slower
(mu = 1/2), and a kernel-started perturbation reads a transiently larger
exponent that converges.
experiments/OD/D3/grade.json K1, K1c, K2, K3. Witness: gap 0.82 to 1.25 in 64 of 64 starts at mu = 2 with the classical exponent at its predicted 1.965; control equal within 0.0072; kernel starts larger on the first window in 83 to 98 percent of starts with median excess +1.5 to +3.5 falling to 0.09 to 0.17 by T = 20; generic starts under every projection within 0.035 of the classical exponent at T = 20. Absorbed by: .
Machine checked
lean/DataMiningAsObservation/ReadOperator.lean,
theorems rank_one_reads_one_direction,
readOp_mulVec, quad_readOp,
quad_readOp_nonneg, readOp_mulVec_eq_zero_iff,
readOp_diag, readOp_offdiag,
readOp_symm, readOp_neg,
affine_const_along_nuisance, readOp_affine,
readOp_sqLength_basis, at observation-data-mining f3914f0;
what the check covers is stated in the book’s appendix
C.
Used in
Data Mining as Observation chapters 1, 2, 3, 6, 9, 11, 12, 13.
Related
read subspace; read operator; quotient; identity reader.
See also
Book equations stated beside the entry’s terms, not defining it: 0.12a, 0.12b.
Ledger rows that cite the entry’s records without naming it: NEG-6, NEG-7.
Sources-table rows that share a record with the entry without naming it: chapter 2 section 2.2, chapter 6 section 6.2, chapter 11 section 11.7.
Status
Generated 2026-09-10 by encyclopedia/generate.py; book
at observation-data-mining f3914f0; the commit of every record is listed
in the encyclopedia’s provenance.