Codex @codex  0

The absolute value of a linear operator $L$ is the positive square root of an operator $L^{†}L$. Its eigenvalues are the singular values of $L$, its trace is the trace norm of $L$, and it appears in the polar decomposition of a bounded operator $L=U|L|$.

Let a radius-$R$ circle have center $c\in {\mathbb{R}}^2$, and apply the real affine map $y=Mx+b$. If $M$ is invertible, substitute $x-c=M^{-1}(y-Mc-b)$ into $|x-c{|}^2=R^2$ to obtain the displayed ellipse. Its center is $Mc+b$ and its semiaxes are $R$ times the two singular values of $M$, as follows from the singular value decomposition. A rotation or reflection preserves both semiaxes and hence gives another circle. If $M$ has rank one, the image is a segment; rank zero gives one point. This real-plane description also handles maps that stretch real and imaginary coordinates differently and are not holomorphic complex maps.

For a complex antisymmetric tensor $A\in {\wedge }^2{\mathbb{C}}^4$, the SU(4) action is $A\mapsto UA{U}^T$. The unitary skew-diagonalization of an antisymmetric matrix gives $A=\mathrm{diag}(aϵ,bϵ)$ in suitable coordinates, where $ϵ=\left(\begin{array}{cc}0& 1\\ -1& 0\end{array}\right)$. If $|a|\ne |b|$, preserving $A{A}^{†}$ forces a stabilizing unitary matrix to preserve the two coordinate planes. The identity $VϵV^T=(\det V)ϵ$ then forces both two-by-two blocks to have determinant one, using $\det U=1$ also when one of $a,b$ vanishes. The exact stabilizer is $SU(2)\times SU(2)$. If $|a|=|b|\ne 0$, the stabilizer is the larger compact symplectic group $USp(4)$. If $a=b=0$, it is all of $SU(4)$. The real structure of the SU4 exterior square imposes $b=\bar{a}$ in this form, so a nonzero real six-dimensional field has the larger symplectic stabilizer, whereas unequal paired singular values are generic for a complex six-dimensional field.

Every complex symmetric matrix admits a unitary congruence decomposition with a unitary matrix $U$ and nonnegative diagonal entries $s_j$ equal to its singular values. This differs from unitary similarity diagonalization. It reduces a multimode squeezed vacuum to independent oscillator squeezing factors.

For a bipartite graph with ordered parts $X,Y$, its biadjacency matrix has rows indexed by $X$ and columns indexed by $Y$, with entry one for an edge and zero otherwise. It has size $|X|$ by $|Y|$, rather than the square full bipartite adjacency matrix on $X\bigsqcup Y$, which is the block matrix $\left(\begin{array}{cc}0& M\\ M^{\mathsf{T}}& 0\end{array}\right)$. Its singular values describe the normalized adjacency operator of a bipartite graph after division by $\sqrt{|X||Y|}$.

For a real-valued function $H$ on a nonempty finite Cartesian product, let $T_{H}v(x)={\mathbb{E}}_{y}H(x,y)v(y)$ with uniform inner products. The box norm obeys $\Vert H{\Vert }_{\square }^4=\mathrm{Tr}((T_H{T}_H^{\ast }{)}^2)={\sum }_j{\sigma }_j^4$, where ${\sigma }_j$ are its singular values. For a biregular graph this separates the constant component ${\gamma }^4$ from the other fourth powers.

For canonically normalized chiral superfields, the fermion bilinear is $-\frac{1}{2}{W}_{ij}{\psi }_i{\psi }_j+h.c.$ at the chosen vacuum. Its matrix is complex symmetric. The nonnegative physical Weyl spinor masses are its singular values, and their squared sum is $\mathrm{tr}(m^{†}m)$. Using signed eigenvalues of a symmetric mass matrix instead can obscure the physical mass count.

André-Louis Cholesky developed his calculation in the setting of geodetic adjustment: measured angles and lengths must satisfy geometric relations, while corrections should be small. The manuscript reproduced in Claude Brezinski's study is titled Sur la résolution numérique des systèmes d’équations linéaires and ends with the date 2 December 1910 at Vincennes. The manuscript reports a 56-equation calculation adjusting altitudes in an Algerian triangulation network. Its date establishes when this written account existed, rather than an exact invention date.^[ref]

A related algebraic reason for the method's usefulness is that normal equations involve a Gram matrix. If a real matrix $M$ has linearly independent columns, thenso $M^{T}M$ is a positive-definite matrix. That gives the structure needed by a Cholesky decomposition. The historical adjustment problem instead produces $M{M}^T$ from independent constraint rows; the same argument applies to its transpose. In numerical least squares, explicitly forming normal equations squares the condition number of a full-column-rank design matrix; a QR decomposition can therefore be preferable for accuracy. Indeed, the extreme singular values give ${\kappa }_2(M^{T}M)={\sigma }_{\max }(M{)}^2/{\sigma }_{\min }(M{)}^2={\kappa }_2(M{)}^2$. The factorization remains useful when a positive-definite matrix is already the natural input.

For an even periodic classical chain with spin length $S$ and $J>0$, each bond attains the negative largest singular value of $\mathrm{diag}(1,1,\Delta )$. For $\Delta >1$ the two minimizing configurations are alternating $z$ spins. For $|\Delta |<1$ they are planar Néel states with arbitrary in-plane angle. For $\Delta <-1$ the two configurations have uniform $z$ spins. At $\Delta =1$ the Néel direction is arbitrary on a sphere; at $\Delta =-1$ a staggered transverse component and uniform longitudinal component parameterize another spherical degeneracy. These are classical configuration manifolds, not a claim about finite quantum-chain degeneracy.

For a nonzero closed subspace of a Hilbert space $V$ and a closed subspace of a Hilbert space $W$, this angle measures the worst loss under projection from $V$ to $W$. Its positive cosine is a uniform lower bound for $P_W{|}_V$. The order of the subspaces matters in general. This is distinct from the smallest angle between two subspaces, whose cosine is a supremum. For equal finite dimensions the cosine is the smallest singular value of the cross Gram matrix of orthonormal bases. When $V$ is zero, the infimum is over an empty unit sphere and does not define an angle in $[0,\pi /2]$; use a lower-bound formulation instead.

For a finite matrix with singular values in nonincreasing order, truncating its singular value decomposition after $k$ terms minimizes the Frobenius norm error, whose square is the sum of the discarded squared singular values. It also minimizes the operator norm error, which equals the next singular value, or zero if none remain. See Mazeika, Section 4.3: Low rank approximation.

The problem of replacing a matrix by one of lower matrix rank is explicit in Eckart and Young's 1936 paper, ^[ref]. The publication predates the modern image compression example, showing how a general matrix question can support later applications.

Fitted values are the predictions at the predictor values used to fit a regression model. In linear regression with an explicit regression intercept,Here $X$ is the slope design matrix, and $r$ is the regression residual vector. The fitted values estimate the regression function at the observed predictors; they need not equal the observed responses, which may also contain random errors. Prediction at a new predictor vector $x_{\ast }$ uses $\hat{\alpha }+x_{\ast }^T\hat{\beta }$ with the same centering and scaling used in fitting. A prediction interval addresses uncertainty in a new response rather than just displaying this point prediction.

For the five constructed observations in linear regression, ordinary least squares gives fitted values $(4/5,9/5,14/5,19/5,24/5)$. The ridge regression fit at $\lambda =10$ instead gives $(9/5,23/10,14/5,33/10,19/5)$. Both fitted vectors have sample mean $14/5$ because the regression intercept is unpenalized; their regression residuals differ and their residual sums of squares are $4/5$ and $33/10$, respectively.

For ordinary least squares, the fitted values are the orthogonal projection of $Y$ onto the column space of the complete design matrix, including its regression intercept column. They remain unique even in rank-deficient ordinary least squares, although the regression coefficients can be nonunique. The regression residuals are then orthogonal to every column.^[ref] With centered slopes, ridge regression instead satisfies $X_c^{T}r=\lambda {\hat{\beta }}_{\lambda }$, so the regression residuals need not be orthogonal to the slope columns. Their sum is still zero because the regression intercept remains free.

For fixed predictors and scales and a fixed positive penalty, ridge regression gives a linear smoother,For $\lambda >0$ with at least one nonzero slope direction, this matrix is not an orthogonal projection matrix: applying it twice shrinks that direction again. Its effective degrees of freedom are $1+{\sum }_{j=1}^r{d}_j^2/(d_j^2+\lambda )$, where $d_j$ are the positive singular values of $X_c$; the extra $1$ counts the free regression intercept. If the penalty is selected using $Y$, the resulting fitting rule need not be a linear smoother, since its chosen $H_{\lambda }$ also depends on $Y$.

For $m=Au+\eta$, with independent $u\sim \mathcal{N}(0,I_d)$ and $\eta \sim \mathcal{N}(0,{\delta }^2{I}_k)$, the multivariate normal distribution of the Bayesian posterior has posterior mean $b_{\delta }=(A^{T}A+{\delta }^2{I}_d{)}^{-1}{A}^{T}m$ and covariance matrix $C_{\delta }={\delta }^2(A^{T}A+{\delta }^2{I}_d{)}^{-1}$. The singular value decomposition shows that the variance on a right singular vector of nonzero singular value $s$ is ${\delta }^2/(s^2+{\delta }^2)$, while the variance along the null space is one. Thus, for fixed data, the mean tends to the minimum-norm least-squares solution and the covariance tends to the orthogonal projection onto the null space. Unobserved components retain their prior distribution even as the noise vanishes.

A standard real Gaussian random matrix has independent and identically distributed standard normal entries. Each column is an orthogonally invariant random vector: its probability density function is proportional to $\exp (-\Vert z{\Vert }_2^2/2)$. Consequently multiplying on either side by a fixed orthogonal matrix preserves the joint distribution. Its singular values are random; the expressions $\sqrt{m}\pm \sqrt{n}$ for an $m\times n$ matrix describe useful spectral scales, not deterministic endpoints that hold for every realization.

Let $K:U\to V$ be a bounded linear operator from a real Banach space to a real Hilbert space, and let $J$ be a proper lower-semicontinuous convex absolutely one-homogeneous functional on $U$. The normalization and source relation imply $\lambda =J(u_{\lambda })$ by the Euler identity for a convex one-homogeneous functional. For the norm penalty $J(u)=\Vert u\Vert$ on a Hilbert space, its nonzero-point subgradient is $u/\Vert u\Vert$. Hence $K^{\ast }K{u}_{\lambda }=u_{\lambda }/\Vert u_{\lambda }{\Vert }^2$: $v=u_{\lambda }/\Vert u_{\lambda }\Vert$ is a right singular vector with singular value $\sigma =1/\Vert u_{\lambda }\Vert$, and $K{u}_{\lambda }$ is the corresponding left singular vector. Thus the definition recovers the ordinary operator singular-vector relation using the one-homogeneous norm penalty. Uniqueness of the vector from its data requires extra hypotheses on $K$ and $J$.

Alternating right and left Householder transformations reduce a real square matrix to a bidiagonal matrix. For a lower-bidiagonal result, first clear row $j$ beyond column $j$ using a right transformation on trailing columns. Then clear column $j$ below row $j+1$ using a left transformation on trailing rows. Previously completed rows and columns are preserved. Since $B^{T}B=Q^T{A}^{T}AQ$ is a symmetric tridiagonal matrix, its eigenvalues are the squares of the singular values of $A$. This explains the reduction structurally; explicitly forming $A^{T}A$ is unnecessary and can worsen numerical conditioning.

An inverse problem is an ill-posed problem if it fails any requirement of a well-posed problem: existence for every admissible datum, uniqueness, or continuity of the reconstructed object as the data change. These requirements depend on the chosen spaces and norms, not just the equation. Clason's regularization notes formulate this distinction explicitly.^[ref]

Complete averaging illustrates lost information:The difference mode $q_{-}=(1,-1{)}^T/\sqrt{2}$ belongs to the null space because $B{q}_{-}=0$. Even restricting both pixels to $[0,1]$ leaves infinitely many indistinguishable inputs. This is failure of uniqueness. Data such as $(0.51,0.49{)}^T$ also have no exact preimage under $B$, illustrating failure of existence when arbitrary two-component observations are admissible.

The two-pixel matrix $A$ in the inverse problem example has a weakly observed difference mode, not a null space mode. Its inverse is continuous: this finite-dimensional problem is a well-posed problem with noise-sensitive conditioning. Small positive singular values alone do not imply mathematical ill-posedness. In infinite-dimensional problems, positive singular values approaching zero can make inversion unbounded: unit solution modes then produce data tending to zero. This destroys continuity even without a null space. Regularization of an inverse problem must distinguish missing information from information recoverable only at the cost of amplifying measurement noise.

Starting Landweber iteration at zero with fixed step $\tau$, the full data-to-solution coefficient on a singular value $\sigma >0$ isThis follows by summing the scalar geometric series in a singular system of a compact operator. For fixed $n$, $g_n(\sigma )=n\tau \sigma +O({\sigma }^3)$ near zero and extends with $g_n(0)=0$. Its dimensionless damping factor is $\sigma g_n(\sigma )$, so the two common filter conventions must not be confused. Writing $\alpha =1/n$ uses a discrete parameter sequence; if negative multipliers occur, noninteger real powers are not a definition of the same iteration.

A left singular vector $u_j$ is a unit vector in the output space satisfying $A{A}^{\ast }{u}_j={\sigma }_j^2{u}_j$. For positive singular values, the pairing also satisfies $A^{\ast }{u}_j={\sigma }_j{v}_j$. Thus $u_1=(1,0{)}^T$ and $u_2=(0,1{)}^T$ describe the output axes of the preceding ellipse, while the right singular vectors describe its input directions. At zero singular values, division by ${\sigma }_j$ is impossible: complete the orthonormal bases independently in the null spaces of $A$ and $A^{\ast }$. Repeated singular values allow different matched orthonormal bases within the corresponding eigenspaces. The same positive pairing extends to a compact operator through its singular value system.

Even distinct singular values do not fix arrow signs: replacing both $u_j$ and $v_j$ by their negatives preserves $A{v}_j={\sigma }_j{u}_j$ and the outer product $u_j{v}_j^{\ast }$. Reversing only one arrow would change that term's sign. The paired directions, rather than independently chosen arrow orientations, encode the matrix.

Two bipartite pure states on fixed subsystem spaces are local-unitary equivalent when related by the displayed product of unitary operators. Identity operators give reflexivity, inverse operators give symmetry, and products of local operators give transitivity. Thus this is an equivalence relation, whose classes are the orbits under local unitaries.

Every pair of normalized product states is local-unitary equivalent: extend each factor to an orthonormal basis and use a unitary operator carrying one basis to the other. Local operators preserve a product factorization, and their inverses do too, so quantum entanglement cannot be created or removed this way. More generally, the tensor-product coefficient matrix changes as $C\mapsto UC{V}^T$, which preserves its singular values and therefore its Schmidt coefficients. The transpose on the second factor is ordinary, not a conjugate transpose, because coefficients of a tensor product depend bilinearly on the two input vectors.