The Observation Theory EncyclopediaFrom TSKAboutBy kindBy chapterBy Lean fileLedgerProvenance

matched bits instrument

DefinitionThe rule that two codes are compared only at the same bit count. At matched bits two codes with equal reconstruction error can still differ for a reader, which is the flip. Chapter 4 section 4.3. Also bit-matched, matched reconstruction.
ExampleTwo 4-bit codes of the same vector are compared; a 4-bit code against a 6-bit code is not a comparison.
BookData Mining as Observation, draft 0.2, commit f3914f0; entry id matched-bits, kind instrument.
Statusmeasures [demonstrated]. Corrections: none recorded.
Defining equationnone
Assumptions and scope
  • The rule that two codes are compared only at the same bit count, since each added bit halves the step and quarters the squared error. At matched bits two codes with equal reconstruction error can still differ for a reader, and the flip is that difference.
  • A comparison that was not bit-matched was refuted by the twenty-five percent codebook confound, and the redesign matched bits before any verdict was read.
Prior artnone recorded
Evidencegeometric-observation/claims/LEDGER.md:63, lean/DataMiningAsObservation/Bit.lean, lean/DataMiningAsObservation/Isotropy.lean
Reviewednot yet reviewed; generated 2026-09-10 from records at the commits on the provenance page.
1 bit, 2 levels2 bits, 4 levels3 bits, 8 levels4 bits, 16 levels
Two codes are compared only at the same bit count.

Equation

none

Conditions

Conditions are curated in entries.toml rather than read from a record.

Ledger

First stated

Chapter 4 section 4.3 of Data Mining as Observation, with the bit-matched redesign in geometric-observation/chapters/ch08_value.md:1-30.

Measurements

none

Failures and corrections

none

Invariance envelope

none declared

Machine checked

lean/DataMiningAsObservation/Bit.lean, theorems levels_succ, levels_add, step_succ, sqError_succ, sqError_antitone, levels_zero, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

lean/DataMiningAsObservation/Isotropy.lean, theorems isotropic_reads_same, isotropic_no_flip, anisotropic_readers_differ, flip_iff_anisotropic, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

Used in

Data Mining as Observation chapters 4, 6, 12.

Related

bit; budget; quantization; control.

See also

Book equations stated beside the entry’s terms, not defining it: 4.2, 4.6, 0.15.

Ledger rows that cite the entry’s records without naming it: GO-2 (pos. half: consumer-projected covariance controls), NEG-4.

Sources-table rows that share a record with the entry without naming it: chapter 1 section 1.4, chapter 3 section 3.3, chapter 4 section 4.3, chapter 4 section 4.6, chapter 6 section 6.2, chapter 8 section 8.2.

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.

← marginmatrix →