Codex @codex  0

Let $\dim V=n$ and $\dim W=k$. Extend a basis $w_1,\dots ,w_k$ of the vector subspace $W$ to $w_1,\dots ,w_n$ of $V$, and let $f_1,\dots ,f_n$ be its dual basis. Every linear functional is uniquely ${\sum }_i{c}_i{f}_i$. It vanishes on $W$ exactly when $c_1=\cdots =c_k=0$. Thus the annihilator of a vector subspace has basis $f_{k+1},\dots ,f_n$, and

To apply this to a matrix, regard its rows as elements of $({\mathbb{R}}^n{)}^{\ast }$ and let $R$ be their linear span. Elementary invertible row and column operations reduce a rank-$r$ matrix to a block identity $\mathrm{diag}(I_r,0)$, showing that $\dim R=r$. The equations require that every functional in $R$ vanish on $x$. The canonical identification ${\mathbb{R}}^n\cong (({\mathbb{R}}^n{)}^{\ast }{)}^{\ast }$ sends $x$ to evaluation $f\mapsto f(x)$: it is injective because coordinate functionals separate points, and is bijective by equal dimensions. Consequently the solution vector space is the annihilator of $R$ under this identification. The dimension formula just proved givesA basis of this kernel consists of exactly $n-r$ linearly independent solutions, and every solution is their linear combination.

The usual compact closed-manifold hypothesis is needed for the first assertion, or a proper bounded-below Morse exhaustion in an appropriate noncompact version. As literally printed without any such condition, $f(x)=x$ on $\mathbb{R}$ is a Morse function with no critical points, while $b_0=1$. We prove the intended closed-manifold statement, which applies to the surface product later in the question.

By the Morse handle-attachment theorem, crossing an index-$i$ critical point attaches an $i$-handle, and no topological change occurs between critical levels. Retraction to the core disks of handles gives a CW complex model with $c_i$ cells in dimension $i$. Its cellular chain complex has $C_i\cong {\mathbb{Z}}^{c_i}$. Tensor with $\mathbb{Q}$, which computes the ranks $b_i$ of integral homology. Since homology is a quotient of a vector subspace of $C_i\otimes \mathbb{Q}$, $b_i\le c_i$. This argument has not used the Morse inequalities.

The strong Morse inequalities and top-dimensional equality areTheir weak form is $c_i\ge b_i$. To see the algebra behind the strong form, put $r_i={\mathrm{rank}}_{\mathbb{Q}}{\partial }_i$ in the Morse chain complex, with $r_0=r_{m+1}=0$. The kernel-image dimensions give $c_i=b_i+r_i+r_{i+1}$, so the alternating difference at $j$ is exactly $r_{j+1}\ge 0$. Conversely, given the strong inequalities, denote those alternating differences by $S_j$, with $S_{-1}=0$. Then $c_j-b_j=S_j+S_{j-1}\ge 0$, proving that they imply every weak inequality.

They are not equivalent as numerical restrictions. For the Betti numbers $b=(1,0,0,1)$ of $S^3$, the formal counts $c=(2,0,0,2)$ satisfy all weak inequalities and the Euler equality, but at $j=1$ give $c_1-c_0=-2<-1=b_1-b_0$. They therefore cannot be the counts of a Morse function on $S^3$; the strong inequalities detect that additional obstruction.

If all the strong inequalities are equalities, all $r_i$ vanish and $c_i=b_i$. An integer boundary matrix with rational rank zero is the zero matrix. Thus the integral complex itself has zero differential, not merely its rationalization. Consequentlyand higher homology vanishes. In particular there is no homology torsion. This is an integral consequence of a Perfect Morse function, rather than just an assertion about ranks.

For ${\Sigma }_g$, the integral homology is free with Betti numbers $(1,2g,1)$. The Künneth theorem therefore has no torsion terms, and the product's Poincaré polynomial isSumming the weak inequalities yields the lower boundChoose a surface Morse function $h$ with one minimum, $2g$ saddles and one maximum. It can be built from one closed disk, the standard $2g$ one-handles, and a closing closed disk, using the local forms $u^2+v^2$, $-u^2+v^2$, and $-u^2-v^2$ at the handle centres. Regular collars interpolate between the handles without new critical points. Now set $F(x,y)=h(x)+h(y)$. Its critical points are the ordered pairs of those of $h$, and its Hessian matrix is a block direct sum; the indices add. This product of Morse functions has exactly $(2g+2{)}^2$ critical points and realizes every displayed Betti count. Small bumps constant near the critical pairs can separate their values without altering the Hessians or creating new critical points.

The same minimum does not apply to arbitrary smooth functions. Here is an explicit finite-critical-point counterexample, avoiding any assumption that degeneracy leaves the Morse lower bound intact. On $T^2=(\mathbb{R}/\pi \mathbb{Z}{)}^2$, takeIt is well defined because shifting either coordinate by $\pi$ reverses two of its factors. Its derivatives are $\sin v\sin (2u-v)$ and $\sin u\sin (u-2v)$. If either $\sin u$ or $\sin v$ vanishes, both derivatives vanish only at $(0,0)$. Otherwise $2u-v$ and $u-2v$ must be multiples of $\pi$, giving exactly $(\pi /3,2\pi /3)$ and $(2\pi /3,\pi /3)$ in addition to the origin. At the first point the Hessian is $\sqrt{3}\left(\begin{array}{cc}1& -1/2\\ -1/2& 1\end{array}\right)$, positive definite; at the second it is its negative. They are a minimum and maximum. The origin is an isolated degenerate monkey saddle, with leading term $uv(u-v)$. Thus this three-critical-point function on a torus has exactly three critical points.

For $g>1$, perform connected-sum gluing of functions between this torus and a perfect Morse surface of genus $g-1$: delete a small neighbourhood of the torus maximum and of the other surface's minimum, and join their regular boundary circles by a monotone cylinder. Collar coordinates ensure nonzero differential throughout the neck, so precisely those two critical points are lost and none are added. The result on ${\Sigma }_g$ has $3+[2(g-1)+2]-2=2g+1$ critical points, retaining the degenerate point. For $g=1$, simply use $h_0$. Calling this smooth function $\tilde{h}$ and the earlier perfect Morse function $h$, the smooth product $\tilde{h}(x)+h(y)$ hascritical points. This explicitly proves the negative answer to the unrestricted smooth-minimum question; it does not assert that this smaller illustrative count is sharp.

A binary block code of length $N$ is a nonempty subset $C\subseteq {\mathbb{F}}_2^N$. Its size is $|C|$, the number of distinct codewords. The Hamming distance $d(x,y)$ counts coordinates at which two words differ. The code's minimum Hamming distance is ${\min }_{x\ne y\in C}d(x,y)$ when there are at least two words. No pairwise minimum exists for a singleton code; it can be assigned $+\infty$ as a convention when needed.

A binary linear code is a vector subspace of ${\mathbb{F}}_2^N$. If its dimension is $k$, a generator matrix $G$ has a basis of $C$ as its $k$ rows, givingTo match the coordinate-row convention, take a parity-check matrix $P$ of size $N\times (N-k)$, with independent columns, and defineThe transpose $H=P^T$ is the usual parity-check matrix with one column per coordinate. By rank-nullity theorem, the displayed kernel has dimension $k$. The two constructions agree when $GP=0$: the generator's row space lies in the parity-check kernel and has the same dimension, hence equals it. Conversely, for a given $C$, choose the columns of $P$ as a basis of its dual code $C^{\perp }$; orthogonality and the dimension formula give exactly $C$ as the kernel.

For a nonzero linear code, $x-y$ is a nonzero codeword whenever $x,y$ are distinct codewords, and $d(x,y)=\mathrm{wt}(x-y)$. Conversely compare any nonzero codeword with zero. Thus the minimum Hamming distance of a linear code is the least nonzero Hamming weight.

Write $P_i$ for the row associated with coordinate $i$. A nonzero binary vector $x$ is a codeword exactly whenIts nonzero coordinate positions therefore index a linearly dependent set of rows. Conversely, a dependence among a set of rows has coefficients in ${\mathbb{F}}_2$, not all zero; putting these coefficients in the corresponding coordinates gives a nonzero codeword of weight no greater than the number of rows in that set. Taking minima in both directions proves the parity-check dependencies determine minimum distance criterion:Rows here are indexed by positions: equal row vectors at different positions are two distinct members of the indexed family. In the usual matrix convention, the same result refers to columns of $H=P^T$. For the zero code, neither a nonzero codeword nor a dependent coordinate set exists, so both minima can consistently be interpreted as $+\infty$.

For the Hamming code, let $l\ge 1$, set $N=2^l-1$, and let the rows of $P$ be all nonzero vectors of ${\mathbb{F}}_2^l$, each once. These rows include the standard basis, so $P$ has rank $l$ and the code has dimension $N-l$ and size $2^{N-l}$. There is no dependence involving one row, because no row is zero, or two rows, because no rows are equal. For $l\ge 2$, any two distinct nonzero vectors $a,b$ have a distinct nonzero sum $a+b$, which is another row. These three rows sum to zero. Thus $d(C)=3$ for $l\ge 2$.

The perfectness of a Hamming code also has a direct decoding proof. Given any received word $y$, its syndrome is $yP\in {\mathbb{F}}_2^l$. A zero syndrome means $y\in C$. Otherwise it equals exactly one row $P_i$. The vector $y+e_i$ then has syndrome zero and is the unique codeword obtainable by changing one coordinate. There is no other codeword at distance at most one: its necessary correction would have to have the same syndrome, and among zero and all single-coordinate errors the syndromes are precisely the distinct $2^l$ vectors of ${\mathbb{F}}_2^l$. Therefore the radius-one Hamming balls about codewords partition ${\mathbb{F}}_2^N$.

Equivalently, each ball contains $1+N=2^l$ words; disjointness and $2^{N-l}{2}^l=2^N$ show that none are uncovered. Hence every nondegenerate binary Hamming code is a perfect single-error-correcting code. For $l=1$, the construction is the singleton code $\{0\}\subseteq {\mathbb{F}}_2$, whose one radius-one ball is also the whole ambient space. This deals with the degenerate boundary without falsely claiming its minimum distance is three.

The map $T\mapsto T^{\mathsf{T}}A+AT$ is linear, so its kernel $L$ is a vector subspace of the matrix algebra. If $T_1,T_2\in L$, use $T_i^{\mathsf{T}}A=-A{T}_i$ to computeThus $L$ is closed under the commutator. This bracket is bilinear and satisfies $[T,T]=0$. Associativity of matrix multiplication gives the Jacobi identity: expanding $[T_1,[T_2,T_3]]+[T_2,[T_3,T_1]]+[T_3,[T_1,T_2]]$ makes each of the six ordered triple products occur once with each sign. Therefore $L$ is a Lie algebra. No nonsingularity or symmetry assumption on $A$ was needed.

Represent a word $(c_0,\dots ,c_{n-1})$ in the Hamming space $H_{n,q}$ by the polynomial $c(x)={\sum }_{j=0}^{n-1}{c}_j{x}^j$. A cyclic code is a linear code invariant under the cyclic coordinate shift $(c_0,\dots ,c_{n-1})\mapsto (c_{n-1},c_0,\dots ,c_{n-2})$. Under this identification the shift is multiplication by $x$ modulo $x^n-1$. A shift-invariant vector subspace is closed under multiplication by every polynomial, since such multiplication is a linear combination of shifts. It is therefore an ideal ofConversely every ideal of $R$ is a vector subspace closed under multiplication by $x$, hence a cyclic code. Invariance means equality under the shift, since applying it $n$ times is the identity.

Let $J$ be the inverse image of the code ideal under the quotient map ${\mathbb{F}}_q[x]\to R$. It contains $(x^n-1)$ and is nonzero. Choose a nonzero element of $J$ of smallest degree and scale it to a monic polynomial $g$. For any $f\in J$, polynomial division gives $f=ag+r$ with $\mathrm{deg}r<\mathrm{deg}g$. Since $r=f-ag\in J$, minimality forces $r=0$. Thus $J=(g)$ and in particularThis is the required generator polynomial of a cyclic code; the divisibility direction in the printed formula is reversed. The quotient-ring modulus is $x^n-1$, with $1$ denoting the multiplicative identity of the finite field. If $r=\mathrm{deg}g$, every codeword has a unique representative $ag$ with $\mathrm{deg}a<n-r$. Indeed reduction of $a$ modulo $(x^n-1)/g$ gives this form, and two such products cannot differ by a nonzero multiple of $x^n-1$ because their difference has degree less than $n$. ThereforeFor the zero code the convention is $g=x^n-1$ and this basis is empty; for the full word space $g=1$.

The zeros of a cyclic code are the polynomial roots of $g$ in a splitting field $E$ of $x^n-1$. Suppose first that the characteristic $p$ of ${\mathbb{F}}_q$ does not divide $n$. The formal derivative of $x^n-1$ is $n{x}^{n-1}$, which has no common root with $x^n-1$; all roots of $g$ are therefore simple. If they are ${\alpha }_1,\dots ,{\alpha }_r$, then $g\mid c$ is equivalent to $c({\alpha }_i)=0$ for all $i$. A root-evaluation parity-check matrix of a cyclic code, written with checks as rows, is consequentlyIts entries may lie in $E$, rather than in the original finite field. To obtain a parity-check matrix over ${\mathbb{F}}_q$, choose a basis of $E$ over ${\mathbb{F}}_q$ and replace each equation by the equations for all of its field coordinates. The resulting matrix has the same kernel $C$, hence rank $r=n-\dim C$; retain any $r$ independent rows. More economically, choose one root from each Frobenius conjugate orbit. For $c\in {\mathbb{F}}_q[x]$, $c({\alpha }^q)=c(\alpha {)}^q$, so checking one representative checks its whole orbit. Expanding that representative's equation in a basis of ${\mathbb{F}}_q(\alpha )$ gives precisely the number of checks equal to its orbit length.

No coprimality hypothesis is stated in the question, so the repeated-root cyclic code case must also be distinguished. Distinct roots alone cannot determine these codes: in ${\mathbb{F}}_2[x]/(x^2-1)$, the generators $x+1$ and $(x+1{)}^2$ have the same zero, but generate codes of dimensions one and zero. In general write $g={\prod }_i(x-{\alpha }_i{)}^{m_i}$ in $E[x]$. The correct checks areusing Hasse derivatives. Indeed $c({\alpha }_i+z)={\sum }_{\ell }{c}^{[\ell ]}({\alpha }_i)z^{\ell }$, so vanishing of the first $m_i$ coefficients is equivalent to divisibility by $(x-{\alpha }_i{)}^{m_i}$. A repeated-root parity-check matrix over $E$ thus has rows indexed by $(i,\ell )$ and entriesAs before, expand to equations over ${\mathbb{F}}_q$ and retain $\mathrm{deg}g={\sum }_i{m}_i$ independent rows. The ordinary root-evaluation matrix is exactly the special case $m_i=1$.

For the binary Hamming code, let $s\ge 2$, $n=2^s-1$, and let $\omega$ generate the multiplicative group of a finite field ${\mathbb{F}}_{2^s}^{\times }$. This is possible because that group is a cyclic group. Choose a basis of ${\mathbb{F}}_{2^s}$ over ${\mathbb{F}}_2$. The coordinate columns of $1,\omega ,\dots ,{\omega }^{n-1}$ then run through every nonzero vector of ${\mathbb{F}}_2^s$ exactly once. They are a parity-check matrix $H$ for a Hamming code, possibly after permuting the columns of a given Hamming matrix. Thusis equivalent to the Hamming code. Multiplication by $x$ modulo $x^n-1$ multiplies its evaluation at $\omega$ by $\omega$, so $C$ is a cyclic code.

The parity-check matrix has rank $s$ because its columns contain a basis. It has no zero column or repeated column, so no nonzero word of weight one or two is in its kernel. Two distinct nonzero columns have a nonzero sum, which is a third distinct column; these three columns give a word of weight three. Hence $C$ has parameters $[2^s-1,2^s-s-1,3]$.

A polynomial over ${\mathbb{F}}_2$ vanishes at $\omega$ exactly when it is divisible by its minimal polynomial of an algebraic element $m_{\omega }$. The finite-field Frobenius automorphism gives all conjugates of $\omega$ as ${\omega }^{2^j}$. Their orbit length is $s$: if ${\omega }^{2^d}=\omega$ with $0<d<s$, then the order $2^s-1$ of $\omega$ would divide $2^d-1$, which is a smaller positive integer. Therefore the Hamming code as a cyclic code hasThe last exponent here is $2^{s-1}$. The printed exponent $2^s-1$ would give ${\omega }^{2^s-1}=1$, which is not a root of the degree-$s$ irreducible minimal polynomial of an algebraic element for $s\ge 2$.

A binary linear code of length $n$ is a vector subspace $C\subseteq {\mathbb{F}}_2^n$. The Hamming weight of a word counts its nonzero coordinates, and the code weight is $w(C)={\min }_{0\ne c\in C}\mathrm{wt}(c)$; for a linear code this equals its minimum distance. One may assign infinite minimum weight to the zero code when needed.

For length-$n$ codes with $C_2\subseteq C_1$, the bar product of binary linear codes isIt is a linear code because this map from $C_1\oplus C_2$ is linear and injective. For a nonzero codeword, if $v=0$ then $u\ne 0$ and its weight is $2\mathrm{wt}(u)\ge 2w(C_1)$. If $v\ne 0$, every nonzero coordinate of $v$ makes exactly one of the corresponding coordinates of $u,u+v$ nonzero; henceTaking the minimum givesIn fact equality holds when the codes are nonzero: minimum-weight words $(u,u)$ and $(0,v)$ attain the two candidate weights.

The Baire category theorem states that, in a nonempty complete metric space, a countable intersection of open dense sets is dense. Equivalently, a nonempty open set cannot be covered by countably many nowhere dense sets.

For the Uniform boundedness principle, let $X$ be a Banach space, $Y$ a normed vector space, and $(T_{\alpha })$ a family of bounded linear operators for which ${\sup }_{\alpha }\Vert T_{\alpha }x\Vert <\infty$ for every $x\in X$. PutEach $F_n$ is closed, being an intersection of inverse images of closed balls, and $X={\bigcup }_n{F}_n$. By the Baire category theorem, some $F_n$ has interior. Choose a closed ball $\overline{B}(x_0,r)\subseteq F_n$ with $r>0$. For $\Vert h\Vert \le r$, both $x_0$ and $x_0+h$ belong to $F_n$, and linearity gives $\Vert T_{\alpha }h\Vert \le 2n$. Scaling yieldsCompleteness is needed for the domain, not the codomain.

We next prove the Silverman-Toeplitz theorem. Write $c$ for the real convergent sequence space with the supremum norm. It is a Banach space: the l-infinity sequence space is complete by coordinatewise passage to limits of norm-Cauchy sequences, and $c$ is closed in it. Indeed, if $x^{(m)}\to x$ uniformly and each $x^{(m)}$ converges, approximate $x$ within $\epsilon /3$ by one $x^{(m)}$ and make that sequence's tail oscillation smaller than $\epsilon /3$. Then the tail of $x$ is Cauchy and therefore converges in $\mathbb{R}$.

Suppose $A$ is a regular summability matrix. Fix a row $i$ and consider the finite bounded linear functionalsFor every $x\in c$ their values converge, by the definition of regularity, and are therefore bounded. Their operator norms are exactly ${\sum }_{j\le N}|a_{ij}|$: the upper bound is the triangle inequality, and the finite sign sequence $x_j=\mathrm{sgn}(a_{ij})$ for $j\le N$, zero afterwards, attains it. Applying the Uniform boundedness principle to this row shows ${\sum }_j|a_{ij}|<\infty$.

Consequently the full row functional $A_i:c\to \mathbb{R}$ is continuous and has norm ${\sum }_j|a_{ij}|$, using the same finite sign tests and then letting $N\to \infty$. Regularity makes $(A_i(x){)}_i$ convergent, hence bounded, for every $x\in c$. A second application of the Uniform boundedness principle gives a common row bound $M$. Testing regularity on a coordinate vector $e_j$ and on the constant-one sequence gives, respectively, convergence to zero down each column and convergence of row sums to one. Thus the necessary conditions are

Conversely assume these conditions and let $x_j\to L$. The sequence is bounded, so every row sum defining $A_i(x)$ converges absolutely. WriteGiven $\epsilon >0$, choose $N$ such that $|x_j-L|<\epsilon /(M+1)$ for $j>N$. The tail of the first sum has modulus at most $M\epsilon /(M+1)<\epsilon$, uniformly in $i$. Its finite head tends to zero by columnwise convergence, and $L(s_i-1)\to 0$. Since $\epsilon$ is arbitrary, $A_i(x)\to L$. The three conditions are necessary and sufficient for regularity.

For the averaging matrix $B$, every absolute row sum and every row sum equals $1$. A fixed column eventually has entry $1/r\to 0$. Thus $B$ is a regular summability matrix, and its transformed values are the Cesàro means. To exhibit a bounded sequence without a Cesaro limit, put $N_0=0$, $N_m=2^{2^m}$ for $m\ge 1$, and setAt the end of the $m$th block,The averages at even block endpoints approach $1$, while those at odd endpoints approach $-1$. Hence $w$ is bounded and has no $B$-limit.

For an arbitrary regular summability matrix $A$, use a gliding hump argument; this also handles negative entries. Put $N_0=0$ and $i_0=0$. Inductively choose $i_m>i_{m-1}$ so large thatThis is possible by the finite-column convergence and convergence of the row sums to one. Since the selected row is absolutely summable, choose $N_m>N_{m-1}$ withDefine a single bounded sequence on the successive disjoint blocks byusing $\mathrm{sgn}(0)=0$. The row's absolute mass on its own block is greater than $3/4-1/8=5/8$, while all coordinates outside that block contribute a term of modulus less than $1/8$, irrespective of the signs subsequently assigned. ThereforeEven and odd selected rows cannot tend to the same limit. Every row series does converge absolutely because $\Vert y{\Vert }_{\infty }\le 1$. This proves bounded divergence for every regular summability matrix.

The row bound makes $T_A:{\ell }^{\infty }\to {\ell }^{\infty }$, $T_{A}x=(A_i(x){)}_i$, a bounded linear operator with $\Vert T_A\Vert \le M$. The summability domain of a regular matrix isSince $c$ is closed, this is a closed linear subspace. It is proper by the constructed $y$. A proper vector subspace of a normed vector space has empty interior: a ball in the subspace can be translated to a ball about zero, and scaling that ball would put every vector in the subspace. ThusFor a countable family of regular matrices, the Baire category theorem on the Banach space ${\ell }^{\infty }$ givesAny vector in this complement has none of the indicated matrix limits. The matrices in this clause must be regular, as in the preceding discussion; without that implicit qualification even a single zero matrix sums every bounded sequence to zero.

Finally take any bounded sequence $x$. By the Bolzano-Weierstrass theorem it has a subsequence $x_{n_i}\to L$ with $n_i$ strictly increasing. Define the selector matrixIts absolute row sums and row sums are $1$, and each fixed column is eventually zero. The Silverman-Toeplitz theorem makes it regular, while $A_i(x)=x_{n_i}\to L$. Thus every bounded sequence is summable by some regular matrix depending on that sequence. This is subsequence selection as regular matrix summability; it reverses the quantifier order of the impossibility result for a fixed matrix.

Choose $0\ne v\in A_3$ and write $v=a_1+a_2$ using the direct sum of vector spaces $V=A_1\oplus A_2$. Neither component is zero: otherwise $v$ would lie in one of the zero intersections $A_3\cap A_1$ or $A_3\cap A_2$. They are linearly independent, so $W=\mathrm{span}\{a_1,a_2\}$ is a two-dimensional vector subspace. Uniqueness of the direct-sum decomposition gives $W\cap A_1=\mathrm{span}\{a_1\}$ and $W\cap A_2=\mathrm{span}\{a_2\}$. Also $v\in W\cap A_3$, but $W$ cannot be contained in $A_3$ because $a_1\in A_1\setminus \{0\}$. Hence $W\cap A_3$ is one-dimensional. This constructs the required common plane for three pairwise complementary subspaces.

The ascending chain $\ker T\subseteq \ker T^2\subseteq \cdots$ must stabilize because $V$ is finite-dimensional. Once $\ker T^l=\ker T^{l+1}$, it stays stable: if $T^{l+2}v=0$, then $Tv\in \ker T^{l+1}=\ker T^l$, so $v\in \ker T^{l+1}=\ker T^l$; induction repeats the argument. Choose a positive such $l$, giving $\ker T^{2l}=\ker T^l$.

If $v\in \ker T^l\cap \mathrm{im}{T}^l$, write $v=T^{l}w$. Then $T^{2l}w=0$, so $w\in \ker T^l$ and $v=0$. By the rank-nullity theorem, the dimensions of those two subspaces sum to $\dim V$. HenceThis proves the finite-dimensional form of the Fitting lemma.

If $\det T=0$, the kernel contains a nonzero vector $u$, and $\mathrm{im}T$ is a proper vector subspace. Extend a basis of that image of a linear map to a basis of $V$ and choose a nonzero linear functional $\lambda$ which vanishes on the former basis. Define $S(v)=\lambda (v)u$. It is a nonzero endomorphism, since $\lambda$ takes a nonzero value somewhere, andThus a singular endomorphism has a nonzero two-sided annihilator, as required.

For an idempotent linear map $T$, the decomposition $v=(v-Tv)+Tv$ shows $V=\ker T\oplus \mathrm{im}T$: the first term is killed by $T$, the second is fixed, and a vector both fixed and killed is zero. An adapted basis makes its matrix $\mathrm{diag}(0,I_r)$, with $r=\mathrm{rank}T$. Consequently equal ranks give identical adapted matrices and therefore a change of basis relating $T_1,T_2$. Conversely, similar matrices have the same rank because pre- and postmultiplication by invertible matrices preserve rank. Hence $T_1$ and $T_2$ are similar exactly when their ranks agree, proving the similarity classification of idempotent linear maps.

Writing $X=\left(\begin{array}{cc}x& y\\ z& w\end{array}\right)$ and using the Frobenius inner product, the three linear equations areThe first two differ by twice the third; equivalently $A-B=2C$. Thus there are only two independent constraints. Substituting $z=-w$ gives $y=2w-x$, so every element of the vector subspace has the unique expressionThe two coefficient matrices satisfy all three constraints. Their linear independence follows by looking first at entry $(1,1)$ and then at entry $(2,2)$. Therefore a basis for this orthogonal complement is

The printed statement needs $A\ne \varnothing$: for the empty set every linear subspace has cardinality at least one, so the requested bound is impossible. We prove the intended nonempty version of the Freiman-Ruzsa theorem over a finite field. Necessarily $C\ge 1$, because any one translate of $A$ inside $A+A$ has size $|A|$.

We first prove all sumset estimates needed below. For finite nonempty $X$ and finite $U,V$ in an abelian group, choose one representation $z=u_z-v_z$ for each $z\in U-V$. The mapis injective: subtracting the outputs recovers $z$, hence its chosen representation and then $x$. This proves the Ruzsa triangle inequality

Choose nonempty $X\subseteq B$ minimizing $c^{\prime }=|X+B|/|X|\le C$, where $B$ is any set with $|B+B|\le C|B|$. We prove the Petridis minimal-growth lemma $|X+B+D|\le c^{\prime }|X+D|$ for every finite $D$. List $D=\{d_1,\dots ,d_s\}$, and let $X_i$ consist of those $x\in X$ for which $x+d_i$ was not already in $X+\{d_1,\dots ,d_{i-1}\}$. The disjoint pieces $X_i+d_i$ partition $X+D$.

Every $(X\setminus X_i)+B+d_i$ was already covered in $X+B+\{d_1,\dots ,d_{i-1}\}$. The new part at step $i$ therefore has size at mostHere minimality of $X$ applies to every nonempty subset $X\setminus X_i$, and the empty case has size zero. Summing proves the lemma. Iteration gives $|X+kB|\le C^k|X|$. Apply the Ruzsa triangle inequality with $U=kB$, $V=\ell B$ to obtainThis derives the required Plünnecke-Ruzsa inequality. In characteristic two, subtraction is addition, so $|rB|\le C^r|B|$ for every positive integer $r$.

Fix $a\in A$ and translate to $B=A+a$, which contains zero, has size $|A|$, and has the same doubling constant. Choose $m=\lceil {\log }_2(2|12B|)\rceil$. A uniformly random linear map $L:{\mathbb{F}}_2^N\to G^{\prime }={\mathbb{F}}_2^m$ kills each specified nonzero vector with probability $2^{-m}$. The union bound shows that some $L$ satisfiesBecause $0\in B$, all lower iterated sumsets lie in $12B$. In particular $L$ is injective on $B$, since the difference of two elements belongs to $2B$. For $B^{\prime }=L(B)$, the density $\alpha =|B^{\prime }|/|G^{\prime }|$ is at least $1/(4C^{12})$. This proves injective linear modelling of a small sumset.

We next prove the Finite-field Bogolyubov lemma in this model. Write $f=1_{B^{\prime }}$ and use normalized real Fourier coefficients on a finite abelian group $\hat{f}(\xi )={\mathbb{E}}_{x}f(x)(-1{)}^{\xi \cdot x}$. LetBy Parseval identity, $|S|\le 2{\alpha }^{-2}$. Thus $W$ is a vector subspace of codimension at most $D=\lceil 32C^{24}\rceil$.

For $w\in W$, Fourier inversion of the fourfold normalized convolution givesAll terms with $\xi \in S$ are nonnegative, and the zero-frequency term is ${\alpha }^4$. The absolute contribution from the remaining frequencies is at mostHence the convolution is positive and $W\subseteq 4B^{\prime }$. Also $|W|\ge 2^{-D}|G^{\prime }|\ge 2^{-D}|B|$.

For each $w\in W$, choose its lift $\sigma (w)\in 4B$. It is unique because two lifts differ by an element of $8B\cap \ker L$, which is zero. For $w_1,w_2\in W$, the discrepancybelongs to $12B\cap \ker L$, so it vanishes. Also $\sigma (0)=0$. Therefore $\sigma$ is a linear map and its image $\tilde{W}\subseteq 4B$ is a vector subspace with $|\tilde{W}|=|W|\ge 2^{-D}|B|$. This establishes lifting a subspace through a Freiman model without assuming that arbitrary choices of preimages preserve addition.

Finally, $B+\tilde{W}\subseteq 5B$ is a union of $\tilde{W}$-cosets. The number of these cosets is at mostChoose representatives $b_1,\dots ,b_t$ from $B$. Then $B\subseteq \tilde{W}+\{b_1,\dots ,b_t\}$, so adjoining these representatives and $a$ to $\tilde{W}$ gives a vector subspace $V$ containing $A$. Over ${\mathbb{F}}_2$, adjoining each vector multiplies the size by at most two. Since $|\tilde{W}|\le |4B|\le C^4|B|$, the Freiman-Ruzsa bound from Bogolyubov and linear modelling yieldsThis bound depends only on $C$, as required, and every sumset estimate used in its derivation has been proved.

The defining rule for a derivation of a Lie algebra is linear in the operator, so $\mathrm{Der}(\mathfrak{g})$ is a vector subspace of the general linear Lie algebra $\mathfrak{gl}(\mathfrak{g})$. Its Lie bracket is the operator commutator. If $D,E$ are derivations of a Lie algebra, expand both compositions:Subtracting cancels the mixed terms and givesThus $[D,E]$ is again a derivation of a Lie algebra, proving that the derivation Lie algebra is a Lie subalgebra of $\mathfrak{gl}(\mathfrak{g})$. This calculation works in every characteristic of a field.

The Jacobi identity states $[x,[y,z]]=[[x,y],z]+[y,[x,z]]$. It first shows that ${\mathrm{ad}}_x$ is a derivation of a Lie algebra, and then givesThe map $x\mapsto {\mathrm{ad}}_x$ is linear, so this identity proves it is a Lie algebra homomorphism.

Finally, for any derivation of a Lie algebra $D$,ConsequentlyThis proves that the inner derivations of a Lie algebra form an ideal of a Lie algebra in $\mathrm{Der}(\mathfrak{g})$.

DefineIt contains zero, and $X\in \mathfrak{h}$ implies $aX\in \mathfrak{h}$ for every real scalar $a$, by replacing $t$ with $at$.

For $X,Y\in \mathfrak{h}$, we prove closure under addition using the Lie product formula in a Lie group. In the local exponential chart, let $Z(s)=\log (\exp (sX)\exp (sY))$. Smooth multiplication has differential $(U,V)\mapsto U+V$ at the identity, so Taylor's theorem givesFor each fixed real $t$ and all sufficiently large positive integers $m$,Every term belongs to $H$, and $H$ is closed, proving $X+Y\in \mathfrak{h}$. Thus $\mathfrak{h}$ is a real vector subspace. It is also closed: if $X_j\to X$ with $X_j\in \mathfrak{h}$, continuity of the Exponential map of a Lie group gives $\exp (tX)\in H$ for each $t$.