-
Definition · Subject of the article
✕Not formalized
Dimensionality reduction, or dimension reduction, is the transformation of data from a high-dimensional space into a low-dimensional space so that the low-dimensional representation retains some meaningful properties of the original data, ideally close to its intrinsic dimension.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 4 queries and 2 candidates examined (receipt 420fb907d99c). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Definition · main statement · Feature selection
✕Not formalized
The process of feature selection aims to find a suitable subset of the input variables (features, or attributes) for the task at hand.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt c0fc454514e3). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Definition · main statement · Feature projection
✕Not formalized
Feature projection (also called feature extraction) transforms the data from the high-dimensional space to a space of fewer dimensions.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt f8f11a97799b). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Definition · main statement · Feature projection › Principal component analysis (PCA)
✕Not formalized
The main linear technique for dimensionality reduction, principal component analysis, performs a linear mapping of the data to a lower-dimensional space in such a way that the variance of the data in the low-dimensional representation is maximized.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt e8a800377926). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Related, but states something else
IsSelfAdjoint.hasEigenvector_of_isMaxOn related
theorem IsSelfAdjoint.hasEigenvector_of_isMaxOn {𝕜 : Type*} [RCLike 𝕜] {E : Type*} [NormedAddCommGroup E] [InnerProductSpace 𝕜 E] [CompleteSpace E] {T : E →L[𝕜] E} : IsSelfAdjoint T → ∀ {x₀ : E}, x₀ ≠ 0 → IsMaxOn T.reApplyInnerSelf (Metric.sphere 0 ‖x₀‖) x₀ → Module.End.HasEigenvector (↑T) (↑(⨆ x, T.rayleighQuotient ↑x)) x₀
Mathlib.Analysis.InnerProductSpace.Rayleigh · lines 276–290
How it differs: This is the variational fact behind the passage's maximisation, but the passage defines a map of the data into fewer dimensions, which this declaration never mentions.
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 1 for this passage, high confidence. - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: it survived. - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
3dfb8e2. - Location key
ac5611b43ed6 · pair key 3ef7d6c7ead3.
Sign in to tag this passage
-
Theorem · supporting · Feature projection › Principal component analysis (PCA)
✕Not formalized
The eigenvectors that correspond to the largest eigenvalues (the principal components) can now be used to reconstruct a large fraction of the variance of the original data.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 4 queries and 2 candidates examined (receipt b6ee15a1b9ea). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Related, but states something else
LinearMap.IsSymmetric.sum_re_inner_apply_self_le_sum_eigenvalues related
theorem LinearMap.IsSymmetric.sum_re_inner_apply_self_le_sum_eigenvalues {𝕜 : Type*} {E : Type*} [RCLike 𝕜] [NormedAddCommGroup E] [InnerProductSpace 𝕜 E] [FiniteDimensional 𝕜 E] {T : E →ₗ[𝕜] E} {n : ℕ} (hT : T.IsSymmetric) (hn : Module.finrank 𝕜 E = n) {ι : Type*} [Fintype ι] {w : ι → E} : Orthonormal 𝕜 w → ∑ i, RCLike.re (Inner.inner 𝕜 (T (w i)) (w i)) ≤ ∑ j with ↑j < Fintype.card ι, hT.eigenvalues hn j
TauCeti.Analysis.InnerProductSpace.KyFan · lines 343–376
How it differs: Ky Fan's inequality bounds what any orthonormal family of directions can capture by the sum of the leading eigenvalues; it is an upper bound and claims nothing about how much of the variance the leading eigenvectors actually recover.
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 1 for this passage, high confidence. - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC). - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
83245ab. - Location key
29c3f681c0cf · pair key 2d59c87ef65c.
Sign in to tag this passage
-
Theorem · supporting · Feature projection › Principal component analysis (PCA)
✕Not formalized
The original space (with dimension of the number of points) has been reduced (with data loss, but hopefully retaining the most important variance) to the space spanned by a few eigenvectors. [citation needed]
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt b46dc9af6543). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Related, but states something else
ContinuousLinearMap.exists_rank_le_nnnorm_sub_eq_singularValue related
theorem ContinuousLinearMap.exists_rank_le_nnnorm_sub_eq_singularValue {𝕜 : Type*} {E : Type*} {F : Type*} [RCLike 𝕜] [NormedAddCommGroup E] [InnerProductSpace 𝕜 E] [FiniteDimensional 𝕜 E] [NormedAddCommGroup F] [InnerProductSpace 𝕜 F] (T : E →L[𝕜] F) (n : ℕ) : ∃ R, (↑R).rank ≤ ↑n ∧ ‖T - R‖₊ = T.singularValue n
TauCeti.Analysis.OperatorIdeal.ApproximationNumber.Attainment · lines 100–129
How it differs: This is the nearest formal analogue of trading dimension for error, as it realises the best approximation by a map of bounded rank, but it mentions no eigenvector span and no data set.
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 1 for this passage, medium confidence. - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC). - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
83245ab. - Location key
7d07493f6da8 · pair key 4640b8387e21.
Sign in to tag this passage
-
Definition · main statement · Feature projection › Non-negative matrix factorization (NMF)
✕Not formalized
NMF decomposes a non-negative matrix to the product of two non-negative ones, which has been a promising tool in fields where only non-negative signals exist,[7][8] such as astronomy.[9][10]
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 4 queries and 2 candidates examined (receipt e4dbb5379806). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Construction · main statement · Feature projection › Kernel PCA
✕Not formalized
Principal component analysis can be employed in a nonlinear way by means of the kernel trick.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt 32bb6c78b2d6). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Related, but states something else
Matrix.PosSemidef.inner_kolmogorovFeature related
theorem Matrix.PosSemidef.inner_kolmogorovFeature {𝕜 : Type u} [RCLike 𝕜] {α : Type v} {K : α → α → 𝕜} (hK : Matrix.PosSemidef K) (a b : α) : Inner.inner 𝕜 (hK.kolmogorovFeature a) (hK.kolmogorovFeature b) = K a b
TauCeti.Analysis.PositiveDefinite.Kernel.Kolmogorov · lines 143–152
How it differs: This is the mechanism the passage alludes to, an inner product in a feature space computed by the kernel, but no component analysis is carried out with it anywhere in the universe.
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 1 for this passage, high confidence. - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: it survived. - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
83245ab. - Location key
9b6f4093442b · pair key a6eaa46f6c88.
Sign in to tag this passage
-
Theorem · supporting · Feature projection › Kernel PCA
✕Not formalized
The resulting technique is capable of constructing nonlinear mappings that maximize the variance in the data.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 5 queries and 3 candidates examined (receipt a0264d682afb). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Sign in to tag this passage
-
Definition · main statement · Feature projection › Kernel PCA
✕Not formalized
The resulting technique is called kernel PCA.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 4 queries and 2 candidates examined (receipt 28593ac9bfec). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Construction · supporting · Feature projection › Graph-based kernel PCA
✕Not formalized
Other prominent nonlinear techniques include manifold learning techniques such as Isomap, locally linear embedding (LLE),[13] Hessian LLE, Laplacian eigenmaps, and methods based on tangent space analysis.[14] These techniques assume that the high-dimensional input data lies near a low-dimensional manifold embedded in the ambient space, and construct a low-dimensional representation using a cost function that retains local properties of the data; they can be viewed as defining a graph-based kernel for Kernel PCA.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 5 queries and 2 candidates examined (receipt 201ca37ecc03). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Sign in to tag this passage
-
Theorem · supporting · Feature projection › Graph-based kernel PCA
✕Not formalized
These techniques assume that the high-dimensional input data lies near a low-dimensional manifold embedded in the ambient space, and construct a low-dimensional representation using a cost function that retains local properties of the data; they can be viewed as defining a graph-based kernel for Kernel PCA.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 4 queries and 2 candidates examined (receipt 6171a83a1244). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Definition · main statement · Feature projection › Graph-based kernel PCA
✕Not formalized
The central idea of MVU is to exactly preserve all pairwise distances between nearest neighbors (in the inner product space) while maximizing the distances between points that are not nearest neighbors.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 5 queries and 2 candidates examined (receipt c33f1ca0bfd6). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Construction · supporting · Feature projection › Graph-based kernel PCA
✕Not formalized
An alternative approach to neighborhood preservation is through the minimization of a cost function that measures differences between distances in the input and output spaces.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 4 queries and 2 candidates examined (receipt 6b43b6c48f6e). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Sign in to tag this passage
-
Theorem · supporting · Feature projection › Graph-based kernel PCA
✕Not formalized
Important examples of such techniques include: classical multidimensional scaling, which is identical to PCA; Isomap, which uses geodesic distances in the data space; diffusion maps, which use diffusion distances in the data space; t-distributed stochastic neighbor embedding (t-SNE), which minimizes the divergence between distributions over pairs of points; and curvilinear component analysis.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 4 queries and 2 candidates examined (receipt 5c2733c23942). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Sign in to tag this passage
-
Definition · supporting · Feature projection › Graph-based kernel PCA
✕Not formalized
Important examples of such techniques include: classical multidimensional scaling, which is identical to PCA; Isomap, which uses geodesic distances in the data space; diffusion maps, which use diffusion distances in the data space; t-distributed stochastic neighbor embedding (t-SNE), which minimizes the divergence between distributions over pairs of points; and curvilinear component analysis.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 4 queries and 2 candidates examined (receipt 959c4169387a). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Sign in to tag this passage
-
Definition · supporting · Feature projection › Graph-based kernel PCA
✕Not formalized
Important examples of such techniques include: classical multidimensional scaling, which is identical to PCA; Isomap, which uses geodesic distances in the data space; diffusion maps, which use diffusion distances in the data space; t-distributed stochastic neighbor embedding (t-SNE), which minimizes the divergence between distributions over pairs of points; and curvilinear component analysis.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 2 candidates examined (receipt 4091b9fc38de). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Sign in to tag this passage
-
Definition · supporting · Feature projection › Graph-based kernel PCA
✕Not formalized
Important examples of such techniques include: classical multidimensional scaling, which is identical to PCA; Isomap, which uses geodesic distances in the data space; diffusion maps, which use diffusion distances in the data space; t-distributed stochastic neighbor embedding (t-SNE), which minimizes the divergence between distributions over pairs of points; and curvilinear component analysis.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt 74d55a860b25). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Related, but states something else
InformationTheory.klDiv related
def InformationTheory.klDiv {α : Type*} {mα : MeasurableSpace α} : MeasureTheory.Measure α → MeasureTheory.Measure α → ENNReal
Mathlib.InformationTheory.KullbackLeibler.Basic · lines 56–59
How it differs: defines the divergence of one measure from another on a single measurable space; it introduces no embedding and no pair-similarity distributions derived from one
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 1 for this passage, high confidence. - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC). - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
3dfb8e2. - Location key
bb466a706a31 · pair key 2e8ebd6fe922.
Sign in to tag this passage
-
Definition · supporting · Feature projection › Graph-based kernel PCA
✕Not formalized
A different approach to nonlinear dimensionality reduction is through the use of autoencoders, a special kind of feedforward neural networks with a bottleneck hidden layer.[16]
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt 6904c3b31b7a). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Related, but states something else
MetaMathlibExt.universal_approximation related
theorem MetaMathlibExt.universal_approximation {n : ℕ} {K : Set (EuclideanSpace ℝ (Fin n))} : IsCompact K → ∀ {σ : ℝ → ℝ}, Continuous σ → (¬∃ p, ∀ (x : ℝ), σ x = Polynomial.eval x p) → ∀ {f : EuclideanSpace ℝ (Fin n) → ℝ}, ContinuousOn f K → ∀ {ε : ℝ}, 0 < ε → ∃ m c W b, ∀ x ∈ K, |f x - ∑ i, c i * σ (Inner.inner ℝ (W i) x + b i)| < ε
MathlibExt.Analysis.Approximation.UniversalApproximation · line 1067
How it differs: states that networks with one hidden layer approximate continuous functions uniformly on compacta; it fixes no narrow middle layer and no reduced representation of the input
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 1 for this passage, high confidence. - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: it survived. - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
64c2773. - Location key
a723c8075016 · pair key 9dc96f3b4c48.
Sign in to tag this passage
-
Definition · main statement · Feature projection › Linear discriminant analysis (LDA)
✕Not formalized
Linear discriminant analysis (LDA) is a generalization of Fisher's linear discriminant, a method used in statistics, pattern recognition, and machine learning to find a linear combination of features that characterizes or separates two or more classes of objects or events.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 2 candidates examined (receipt 2b076784cc07). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Definition · main statement · Feature projection › Generalized discriminant analysis (GDA)
✕Not formalized
GDA deals with nonlinear discriminant analysis using kernel function operator.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt 2793d079cafa). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Related, but states something else
Matrix.PosSemidef.kolmogorovFeature related
def Matrix.PosSemidef.kolmogorovFeature {𝕜 : Type u} [RCLike 𝕜] {α : Type v} {K : α → α → 𝕜} (hK : Matrix.PosSemidef K) : α → hK.KolmogorovSpace
TauCeti.Analysis.PositiveDefinite.Kernel.Kolmogorov · lines 134–138
How it differs: builds the feature map of a positive semidefinite kernel into a Hilbert space; it carries no class labels and no discriminant criterion computed in that space
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 1 for this passage, high confidence. - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: it survived. - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
83245ab. - Location key
69b41f9556eb · pair key ec44bd62fa3d.
Sign in to tag this passage
-
Definition · supporting · Feature projection › Generalized discriminant analysis (GDA)
✕Not formalized
Similar to LDA, the objective of GDA is to find a projection for the features into a lower dimensional space by maximizing the ratio of between-class scatter to within-class scatter.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 2 candidates examined (receipt d0641e662c7f). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Related, but states something else
TauCeti.centeredScatter related
def TauCeti.centeredScatter (𝕜 : Type*) {E : Type*} {ι : Type*} [RCLike 𝕜] [NormedAddCommGroup E] [InnerProductSpace 𝕜 E] [Fintype ι] : (ι → E) → E →L[𝕜] E
TauCeti.Analysis.InnerProductSpace.CenteredScatter · lines 64–65
How it differs: defines the scatter operator of one finite family of vectors about its own mean; it splits no family into classes and optimises nothing
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 1 for this passage, high confidence. - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC). - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
83245ab. - Location key
59e7d8237bef · pair key bb355fa2e680.
IsCoercive.isLeast_rayleighQuotient related
theorem IsCoercive.isLeast_rayleighQuotient {V : Type*} {H : Type*} [NormedAddCommGroup V] [InnerProductSpace ℝ V] [CompleteSpace V] [NormedAddCommGroup H] [InnerProductSpace ℝ H] {B : V →L[ℝ] V →L[ℝ] ℝ} (hB : IsCoercive B) {J : V →L[ℝ] H} : IsCompactOperator ⇑J → (∀ (u v : V), (B u) v = (B v) u) → J ≠ 0 → IsLeast {r | ∃ v, J v ≠ 0 ∧ (B v) v / ‖J v‖ ^ 2 = r} ‖hB.formSolutionOperator J‖⁻¹
TauCeti.Analysis.InnerProductSpace.Variational.Rayleigh · lines 134–147
How it differs: gives the least value of a symmetric coercive form divided by a squared norm; the two quadratic forms of the passage are the scatters of a labelled sample, and the extremum sought is the other one
Details
- Found by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: candidate 2 for this passage, medium confidence. - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC). - Record set
tierA-b1b, exported 1 Nov 2026 by formal-companion d85385a. - Lean code: the declaration's type as Lean prints it, laid out as a header (the corpus keeps no hypothesis names), from the declaration corpus of universe u1 at commit
83245ab. - Location key
9f669f15a3ae · pair key 6367370a2290.
Sign in to tag this passage
-
Construction · main statement · Feature projection › Autoencoder
✕Not formalized
Autoencoders can be used to learn nonlinear dimension reduction functions and codings together with an inverse function from the coding to the original representation.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 1 candidates examined (receipt 27771d7d06b1). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Definition · main statement · Feature projection › t-SNE
✕Not formalized
T-distributed Stochastic Neighbor Embedding (t-SNE) is a nonlinear dimensionality reduction technique useful for the visualization of high-dimensional datasets.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 3 queries and 2 candidates examined (receipt fa3f5828a34a). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Definition · main statement · Feature projection › UMAP
✕Not formalized
Uniform manifold approximation and projection (UMAP) is a nonlinear dimensionality reduction technique.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 5 queries and 2 candidates examined (receipt 46ac83a047b8). - Re-checked by Claude Opus 5 (claude-opus-5) on 1 Nov 2026 (UTC), challenge run
chal-tierA-b1 by jack: upheld.
Sign in to tag this passage
-
Definition · supporting · Feature projection › UMAP
✕Not formalized
Visually, it is similar to t-SNE, but it assumes that the data is uniformly distributed on a locally connected Riemannian manifold and that the Riemannian metric is locally constant or approximately locally constant.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 6 queries and 3 candidates examined (receipt 78b97b2dca33). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Sign in to tag this passage
-
Definition · supporting · Applications
✕Not formalized
A dimensionality reduction technique that is sometimes used in neuroscience is maximally informative dimensions,[24] which finds a lower-dimensional representation of a dataset such that as much information as possible about the original data is preserved.
Show in article · Read it in context on Wikipedia ↗
✕Searched 48 libraries — not found. Which libraries, at which commits (pinned 2026-10-17).
Details
- Searched by Claude Opus 5 (claude-opus-5) on 31 Oct–1 Nov 2026 (UTC), locate run
loc-tierA-b1b by jack: the search receipt lists 5 queries and 2 candidates examined (receipt 4f5c112782c4). - Not re-checked: sampled out of challenge run
chal-tierA-b1 by jack, in which Claude Opus 5 (claude-opus-5) re-checked a sample on 1 Nov 2026 (UTC).
Sign in to tag this passage