The Observation Theory EncyclopediaFrom TSKAboutBy kindBy chapterBy Lean fileLedgerProvenance

outer product concept

DefinitionThe matrix whose entry in row i and column j is the product of a vector's i-th and j-th coordinates, of rank at most one and positive semidefinite for a vector with itself. Chapter 0 section 0.5.
Example(1, 2) with itself gives the matrix with rows (1, 2) and (2, 4), which has rank one.
BookData Mining as Observation, draft 0.2, commit f3914f0; entry id outer-product, kind concept.
Statusno ledger row names this entry. Corrections: none recorded.
Defining equation

Book equation 0.9.

Assumptions and scope
  • The matrix whose entry in row i and column j is the product of the i-th and j-th coordinates of a vector. Its action on any vector is a multiple of the vector it was built from, it has rank at most one, and the outer product of a vector with itself is positive semidefinite.
  • The read operator is the weighted mean of the outer product of the sensitivity with itself, and a linear classifier’s read operator is one outer product scaled by how steep the score is.
Prior artnone recorded
Evidencelean/DataMiningAsObservation/PositiveSemidefinite.lean, lean/DataMiningAsObservation/ReadOperator.lean, lean/DataMiningAsObservation/Rank.lean
Reviewednot yet reviewed; generated 2026-09-10 from records at the commits on the provenance page.
rank one
The entry in row i and column j is the product of the i-th and j-th coordinates.

Equation

Book equation 0.9.

\[P_C=\mathbb E\!\left[g\,g^{\top}\right],\qquad g=\nabla C(x).\]

Book equation 6.1.

\[s(x)=w\cdot x+b,\qquad P_C=\mathbb E\big[\sigma'(s)^{2}\big]\,w\,w^{\top},\qquad \operatorname{rank}P_C=1.\]

Book equation 0.3.

\[\Sigma_{ij}=\mathbb E\big[(x_i-\mu_i)(x_j-\mu_j)\big],\qquad \operatorname{tr}\Sigma=\sum_{i}\Sigma_{ii}.\]

Conditions

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

Ledger

none

First stated

Chapter 0 section 0.5 of Data Mining as Observation.

Measurements

none

Failures and corrections

none

Invariance envelope

none declared

Machine checked

lean/DataMiningAsObservation/PositiveSemidefinite.lean, theorems psd_add, psd_smul, psd_outer, psd_readOp, psd_diag_nonneg, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

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.

lean/DataMiningAsObservation/Rank.lean, theorems rank_le_width, rank_le_height, rank_outer_le_one, rank_mul_le, rank_zero, rank_transpose, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

Used in

Data Mining as Observation primer L, chapters 0, 2, 6.

Related

read operator; rank; positive semidefinite; covariance matrix; sensitivity.

See also

Ledger rows that cite the entry’s records without naming it: OT-7.

Sources-table rows that share a record with the entry without naming it: chapter 2 section 2.2, chapter 6 section 6.1.

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.

← orthonormal basisoutlier →