Glossary: K
Definitions in this project whose names start with K. The whole glossary has the other sections.
A · B · C · D · E · F · G · H · I · J · K · L · M · N · O · P · Q · R · S · T · U · V · W · X · Y · Z · Other
- K abbrev
- The homotopy category Kᵇ(C); boundedness is part of each object. · OAI.LittleFinitistic.BoundedHomotopy.K
- K abbrev
- K^b(proj-R), as a full subcategory with actual bounded f.g. projective representatives. · OAI.LittleFinitistic.FiniteProjectiveHomotopy.K
- K abbrev
- Section 3's K = Kᵇ(proj-B), with homotopy classes of chain maps. · OAI.LittleFinitistic.ProjectiveComplexes.K
- K abbrev
- A small type of precisely the source's probe multiplicity. · OAI.PartitionConsistency.ConcreteSystem.K
- K abbrev
- The logarithmic bound for the number of geometric levels. · OAI.Problem337.RawDensityParameters.K
- K def
- The closed tetrahedron in physical Euclidean coordinates. · OAI.TriangularCovering.K
- kanBaseChangeWeak def
- Weak equivalence after every base change along an actual Kan fibration. This is a property to be proved for cells, not an extra main-theorem input. · OAI.Thomason.Construction.kanBaseChangeWeak
- kanExtensionComparison def
- The actual comparison to the restriction of a larger total space. · OAI.Thomason.Construction.kanExtensionComparison
- KanQuillenPublishedInput structure
- The portion of the published Kan--Quillen theorem required by transfer, with all three classes fixed to their actual definitions. · OAI.Thomason.Construction.KanQuillenPublishedInput
- kanReplacementFactorization def
- The actual horn factorization of the map to the terminal simplicial set. · OAI.Thomason.Construction.kanReplacementFactorization
- kanSimplexRestriction def
- The canonical restriction map on total spaces above a simplex map. · OAI.Thomason.Construction.kanSimplexRestriction
- KaplanskyColumnApproximation def
- The matrix-column consequence of Kaplansky density (Blackadar, I.9.1.3), together with the C-star first isomorphism theorem and quotient norm lifting. The lifting bound has fixed slack 2. · OAI.KadisonSimilarity.KaplanskyColumnApproximation
- kaplanskyFunction def
- The scalar transform used in bounded approximation. · OAI.KadisonSimilarity.kaplanskyFunction
- kaplanskyOffDiagonal def
- The selfadjoint off-diagonal operator associated to a contraction. · OAI.KadisonSimilarity.kaplanskyOffDiagonal
- kaplanskyPreimage def
- The bounded inverse branch, defined by functional calculus. · OAI.KadisonSimilarity.kaplanskyPreimage
- kappa def
- Linear growth is enough for the normalization coding sizes · OAI.FourierLLogL.Dyadic.Thermal.kappa
- kappa def
- The multiplier in equation (15), with its exact rational coefficients. · OAI.PositiveEinsteinFour.Potential.kappa
- kappa def
- The weight in the stationary profile norm
D. · OAI.StationaryNS.kappa - kappa def
- The universal archimedean error mean, before any configuration is chosen. · OAI.WeakPinned.Grid.kappa
- kazhdanSet def
- An explicit finite set in the honest integral elementary subgroup. · OAI.NuclearUltrapower.Elementary.kazhdanSet
- keepComponent def
- Retain the connected component of an edge. · OAI.KaplanskyCounterexample.Planar.Ribbon.keepComponent
- keepHead def
- A recursive subnetwork is used ONCE, with its prefix wire retained. · OAI.ExactQuantumFactoring.BitArithmetic.SortedWords.keepHead
- keepInterval def
- The literal restricted successful matching, retaining an inserted word. · OAI.FKCRT.Inventory.StackMatching.keepInterval
- keptEdges def
- Edges remaining after DELETION, not pinning, of F. · OAI.BufferedIsing.keptEdges
- kernel def
- The degree
-1kernel inker:transverse-expectation; its value at zero is zero. · OAI.BoltzmannNonuniqueness.TransverseAverage.kernel - kernel def
- One row of a dyadic transition table, viewed as an exact rational law. · OAI.ContingencyTables.IntegerPropagation.kernel
- kernel def
- Exact probabilities for the symmetric nearest-neighbour walk on the integers. · OAI.FKCRT.SimpleWalk.kernel
- kernel def
- Restricted Newton kernel;
cis its (positive) normalization. · OAI.HenonLaneEmden.IntervalGeometry.kernel - kernel def
- The nonnegative, degree-normalized squared geometric sum. · OAI.Problem337.GeometricKernel.kernel
- kernel def
- Endpoint correction on the slow initial data. Both decaying factors are kept explicitly, as required in (future-kernel-graph). · OAI.SmoothObstruction.FiniteGreen.Splitting.Terminal.kernel
- kernel def
- Uniform sampling among all allowed nonzero labels. · OAI.SmoothObstruction.SymplecticSampling.kernel
- kernel def
- Exact two-dimensional rescaling
ε⁻² ρ(x/ε). · OAI.TriangularBilliards.SpatialSmoothing.kernel - kernelApply def
- Apply a finite real kernel. Output indices need no finiteness assumption. · OAI.MaxCutGames.Repetition.kernelApply
- kernelApply def
- Apply a finite real kernel. Output indices need no finiteness assumption. · OAI.UniqueGamesTheorem.Repetition.kernelApply
- kernelAverage def
- An actual average over the testing set, in the first kernel variable. · OAI.RegularPointCoordinates.kernelAverage
- kernelChartHomeomorph def
- A homeomorphism over the parameter base between the actual local kernel incidence space and the product with one fixed kernel. · OAI.Kaplansky.SmoothKernel.kernelChartHomeomorph
- kernelCoefficients def
- Collision blocks are sums of the rank-one matrices with that label value. · OAI.Dixmier.kernelCoefficients
- kernelConjugateIso def
- Transport a kernel through isomorphisms of its source and target. · OAI.CoherentDual.kernelConjugateIso
- kernelCoordinateChange def
- The fiber coordinate transition really is projection onto the second fixed kernel after reconstruction in the varying kernel. · OAI.Kaplansky.SmoothKernel.kernelCoordinateChange
- kernelCutMaximum def
- The unnormalized finite real sign-cut maximum. · OAI.JointDickman.kernelCutMaximum
- kernelCutNorm def
- The manuscript's normalization divides the sign maximum by
N. · OAI.JointDickman.kernelCutNorm - KernelDerivative def
- The derivative interface supplied by the inverse-norm calculus. · OAI.Problem356.GeometryJets.KernelDerivative
- kernelDifference def
- Hidden coordinates relative to one base matrix in the same row fiber. · OAI.MaxCutGames.Decoder.AdviceFibers.kernelDifference
- kernelDifference def
- Hidden coordinates relative to one base matrix in the same row fiber. · OAI.UniqueGamesTheorem.Decoder.AdviceFibers.kernelDifference
- kernelEncoding def
- The measurable phase encoding, defined without a measurable enumeration. · OAI.EuclideanFiveColor.PhaseEncoding.kernelEncoding
- kernelEquiv def
- Explicit linear identification of the actual kernels throughout the split neighborhood. Both maps are restrictions of the constructed projections. · OAI.Kaplansky.SmoothKernel.kernelEquiv
- kernelEquiv def
- The bounded determinant kernel has exactly the degree-
< epolynomials as its parameters, including ate = 0. · OAI.MatrixMultiplication.AuxiliarySeparation.DeterminantBounds.kernelEquiv - kernelEquiv def
- The kernel is canonically isomorphic to one polynomial copy. · OAI.MatrixMultiplication.AuxiliarySeparation.DeterminantKernel.kernelEquiv
- kernelEquiv def
- Exactly one freely chosen bit per active position parametrizes the kernel. · OAI.MaxCutGames.Soundness.KernelCardinality.kernelEquiv
- kernelEquiv def
- Identifies the actual linear kernel with the previously counted concrete kernel. · OAI.MaxCutGames.Soundness.PartnerLinear.kernelEquiv
- kernelEquiv def
- Exactly one freely chosen bit per active position parametrizes the kernel. · OAI.UniqueGamesTheorem.Soundness.KernelCardinality.kernelEquiv
- kernelEquiv def
- Identifies the actual linear kernel with the previously counted concrete kernel. · OAI.UniqueGamesTheorem.Soundness.PartnerLinear.kernelEquiv
- kernelForm def
- The finite coefficient pairing of a scalar kernel. · OAI.BoundedHochschild.KadisonRingrose.kernelForm
- kernelHeat def
- The canonical positive-time continuous representative defined by the kernel. · OAI.RegularPointCoordinates.kernelHeat
- kernelHeightCast def
- Transport along equality of total depths only; no change to a kernel. · OAI.DilutedSpinGlass.kernelHeightCast
- kernelIdeal def
- The common character kernel is a Lie ideal, not just a linear subspace. · OAI.PolycyclicRecognition.TriangularAdjoint.kernelIdeal
- kernelIndicator def
- The actual zero-image event for a linear map. · OAI.MaxCutGames.Gadget.ActualGadget.kernelIndicator
- kernelIndicator def
- The actual zero-image event for a linear map. · OAI.UniqueGamesTheorem.Gadget.ActualGadget.kernelIndicator
- kernelMap def
- Its kernel is the alpha-multiple of the twisted upper stalk. · OAI.KLInvariance.ReflectionPairTranslation.kernelMap
- kernelMass def
- The kernel mass controlling every jet, with absolute values for signed kernels. · OAI.ForcedComputation.BoundedSpatialJets.kernelMass
- kernelMeasure def
- The intrinsic ideal Haar measure transported to the exact kernel. · OAI.PolycyclicRecognition.CartanWordModel.MeasuredCone.kernelMeasure
- KernelMinimal def
- Minimality of the full outgoing boundary cover. The assignment formulation avoids dependent casts and is equivalent to the usual kernel containment. · OAI.KLInvariance.MomentGraph.GradedSheaf.KernelMinimal
- kernelMonomial def
- The kernel monomial
(uw-vs) u^(e-1-j) v^j, withj < e. · OAI.MatrixMultiplication.AuxiliarySeparation.DeterminantKernel.kernelMonomial - kernelOfGame def
- A projection game's predicate and actual question distribution determine the conditional kernel used by the collision argument. · OAI.MaxCutGames.Repetition.kernelOfGame
- kernelOfGame def
- A projection game's predicate and actual question distribution determine the conditional kernel used by the collision argument. · OAI.UniqueGamesTheorem.Repetition.kernelOfGame
- kernelPreMap def
- A map of labels bounded for the kernel quadratic form induces a bounded linear map on the finite kernel vectors. · OAI.BoundedHochschild.KadisonRingrose.kernelPreMap
- kernelProjection def
- The kernel projection is an actual measurable representative. · OAI.KadisonKastler.WeakOperatorRepresentative.kernelProjection
- kernelPushforward def
- Applying a finite stochastic kernel, forgetting its input. · OAI.MaxCutGames.Foundations.Repetition.kernelPushforward
- kernelPushforward def
- Applying a finite stochastic kernel, forgetting its input. · OAI.UniqueGamesTheorem.Foundations.Repetition.kernelPushforward
- kernelQuotientHomeomorph def
- The quotient homeomorphism respects the actual weak topologies. · OAI.EilenbergGanea.CubeRealization.kernelQuotientHomeomorph
- kernelRealization def
- The intrinsic ideal group and the actual kernel have the same topology and multiplication. · OAI.PolycyclicRecognition.kernelRealization
- kernelRelativeEntropy def
- A measurable version of conditional relative entropy, for dominated kernels. · OAI.InvariantIsing.kernelRelativeEntropy
- kernelRows def
- A nonzero row of a division-ring kernel expands into d independent ordinary kernel rows after a faithful d-dimensional representation. · OAI.RationalHitting.BlockDivision.kernelRows
- kernels def
- Index zero is the heat term; the other two indices are Cartesian derivatives. · OAI.ForcedComputation.PlaneScalarMild.kernels
- kernelSourceOperator def
- All three raw kernels define bounded source operators on the finite interval. · OAI.ForcedComputation.PlaneScalarMild.kernelSourceOperator
- KernelTower def
- Finite kernels depending on the full preceding path, before tilting. · OAI.DilutedSpinGlass.KernelTower
- kernelTransport def
- Application of a finite transition kernel. · OAI.TwoPointCorrelations.kernelTransport
- kernelVector def
- A bounded kernel vector, obtained by multiplication by
uw-vs. · OAI.MatrixMultiplication.AuxiliarySeparation.DeterminantBasis.kernelVector - kernelVector def
- The actual kernel vectors used at degree
e: there areeof them. · OAI.MatrixMultiplication.AuxiliarySeparation.DeterminantKernel.kernelVector - Key abbrev
- A complete proposed successor, including its source and target state. · OAI.OneWayLiveness.CellNetwork.Key
- keyAnswer def
- The actual verifier labeling, viewed as a function of exact affine keys. · OAI.MaxCutGames.Decoder.TableKeysRestoration.keyAnswer
- keyAnswer def
- The actual verifier labeling, viewed as a function of exact affine keys. · OAI.UniqueGamesTheorem.Decoder.TableKeysRestoration.keyAnswer
- keys def
- Exactly the keys attained by the finite source. · OAI.TalagrandDiscreteConvexity.FiniteFibers.keys
- kids def
- Children from the distinct prime fields, ignoring zero padding and 2. Repeated labels in DIFFERENT node records are not identified. · OAI.ExactQuantumFactoring.BitArithmetic.ChildQueue.kids
- killed def
- The walk killed on first visiting -1, with both endpoints nonnegative. · OAI.FKCRT.SimpleWalk.killed
- killedHeat def
- The killed heat kernel obtained from the proved discrete reflection principle. · OAI.FKCRT.SimpleWalk.killedHeat
- KillsSpheres def
- A ghost map in one geometric degree, including unbased spheres. · OAI.GroupRingTopology.Cellular.KillsSpheres
- Kind inductive
- Echo labels have k as the predecessor iteration number. · OAI.ExactDerandomization.Net.Kind
- Kind inductive
- Exactly the allowed vertex list, including reversal of splitting. The nonnegative numbers include every zero block before normalization. · OAI.IntegralCharacterVarieties.OccurrenceIncidence.VertexTable.Kind
- Kind inductive
- Pairing patterns at a layer, with the inequality omitted in the diagonal case. · OAI.Problem335.Pairings.Kind
- KineticLocalFlow structure
- A constructed local solution and its first variation. This is output of the local ODE theorem, not a geometric input. · OAI.PositiveEinsteinFour.Geometry.KineticLocalFlow
- KineticTest structure
- A smooth test, periodic in the spatial variable and compact in time and velocity. · OAI.BoltzmannNonuniqueness.KineticTest
- kineticZeroDifferential def
- The differential sends a momentum perturbation to its raised velocity; the position perturbation has zero rate at zero momentum. · OAI.PositiveEinsteinFour.Geometry.kineticZeroDifferential
- KirchbergDerivationCriterion def
- Kirchberg's derivation criterion (1996), as stated in Ozawa, An invitation to the similarity problems, Theorem 1.1, with boundedness supplied by Ringrose (1972), Theorem 2, p. 434. · OAI.KadisonSimilarity.KirchbergDerivationCriterion
- KirchbergOInfinityAbsorption def
- Nuclear strong pure infiniteness implies absorption for a separable algebra, including the zero algebra, without a unit or simplicity hypothesis. · OAI.WeakPureInfiniteness.KirchbergOInfinityAbsorption
- kirchbergPath def
- The imaginary direction of this path is unitary conjugation. · OAI.KadisonSimilarity.kirchbergPath
- Kit structure
- Sufficient resources and access for a family of offset trials. · OAI.Fifths.Single.Kit
- Kit structure
- all chip fields within budget · OAI.PowerSaving.RAM.Bench.Kit
- kkDistance def
- Kadison and Kastler's ordinary distance, without matrix amplification. · OAI.KadisonKastler.kkDistance
- klFamilies def
- The polynomial families selected by their equal-parameter normalization. · OAI.KLInvariance.klFamilies
- KLSStatement def
- The full dimension-free statement: one constant works for every dimension, every isotropic log-concave density, and every smooth compactly supported test. · OAI.LeanBlast.KLS.KLSStatement
- kmsNeighborsEquiv def
- Extrinsic submodule neighbors and the finite KMS neighbor predicate agree. · OAI.MaxCutGames.Inverse.MatrixChart.kmsNeighborsEquiv
- kmsNeighborsEquiv def
- Extrinsic submodule neighbors and the finite KMS neighbor predicate agree. · OAI.UniqueGamesTheorem.Inverse.MatrixChart.kmsNeighborsEquiv
- knotPermutation def
- Move the inserted t² label to the last index, leaving M in its original order. This is only a simultaneous change of basis of the actual pencil. · OAI.CriticalHoneycomb.FiniteFunctional.StripAsymptotic.knotPermutation
- KoebeCircleDomainConjecture def
- Paper 047-02, Introduction,
thm:main: unrestricted Koebe uniformization. · OAI.CircleDomainRigidity.KoebeCircleDomainConjecture - koopman def
- The actual Koopman operator, restricted to the full centered subspace. · OAI.ThreeTorus.koopman
- koszulExponent def
- The sign from reordering the odd factors, with positions ordered as in ordinary tensor words. There is no choice of orientation hidden here. · OAI.CoordinateSweeps.SignedTensor.koszulExponent
- kProjectiveCoproductIsColimit def
- An actual coproduct of K-projective cochain complexes with K-projective coproduct object remains a coproduct in the genuine derived category. · OAI.DerivedBar.kProjectiveCoproductIsColimit
- kraus def
- A finite Kraus sum, with the adjoint on the left. · OAI.MUB6.Weiner.kraus
- kronecker def
- Kronecker multiplicity is the complex dimension of Hom, with the diagonal symmetric-group action on the tensor product. · OAI.Saxl.kronecker
- KummerPrimeSiegelWalfisz def
- The raw prime-sum specialization. Cubes and the zero numerator are excluded explicitly; no smooth or structured-sum estimate is assumed. · OAI.CubicFirstMoment.KummerPrimeSiegelWalfisz
- KWOmega abbrev
- The countable product of unit intervals. · OAI.ExactModuli.KWOmega