Every smooth left-invariant vector field on a Lie group is a complete vector field. Translate a local integral curve of a vector field through the identity to every other point. The same positive local existence interval works at every initial point, so no maximal integral curve of a vector field can have a finite endpoint. Uniqueness then makes the curve through the identity a one-parameter subgroup.

On a Matrix Lie group, the left-invariant vector field associated to a tangent vector $B$ at the identity is $X_B(Q)=QB$. Its ambient derivative is $D{X}_B(Q)[H]=HB$. Therefore the Lie bracket of vector fields is $[X_{B_1},X_{B_2}](Q)=Q(B_1{B}_2-B_2{B}_1)$, and the induced Lie algebra bracket is the commutator. This fixes the sign for left invariance; right-invariant fields induce the opposite sign when evaluated with the same matrix identification.

A basis $T_a$ of the Lie algebra of a Lie group gives a global frame of a vector bundle of its tangent bundle by $X_a(g)=(d{L}_g{)}_e{T}_a$. Left translation is a diffeomorphism, so the left-invariant vector fields are independent everywhere. Every Lie group is therefore a parallelizable manifold.

Choose a nonzero element of ${\Lambda }^n{T}_e^{\ast }G$ for an $n$-dimensional Lie group and transport it by inverse left translations. This gives a smooth nowhere-zero left-invariant differential form of top degree, hence an orientation. The identity $L_{(hg{)}^{-1}}{L}_h=L_{g^{-1}}$ proves invariance. The space of invariant top forms is one-dimensional, since evaluation at the identity identifies it with ${\Lambda }^n{T}_e^{\ast }G$.

On a segment starting at residual-score level $t$, an inactive score evolves as $c_j-h{a}_j$ while the active level is $t-h$. Solving $c_j-h{a}_j=\pm (t-h)$ gives distances $(t-c_j)/(1-a_j)$ and $(t+c_j)/(1+a_j)$. The smallest positive admissible distance determines the next entry; reaching level zero ends the algorithm.

Each least angle regression segment starts with all active scores at magnitude $t$. Moving with direction $d_A=G_{AA}^{-1}{s}_A$ changes the score vector to $t{s}_A-(t-\lambda )G_{AA}{d}_A=\lambda s_A$. The first-hit rule bounds every inactive score in magnitude by $\lambda$. This is the key Lasso optimality link.

If every nonzero active coefficient has the sign of its residual score along the least angle regression path, the LAR active correlation invariant and inactive bounds give the Karush-Kuhn-Tucker conditions for the Lasso. A unique Lasso solution then equals the LAR path. Zero coefficients use a subgradient interval; literal sign equality with sign(0)=0 is not appropriate for a newly entering coefficient at a positive knot.

For $0<\tau <2/\Vert A{\Vert }^2$, the Landweber full inverse filter obeysIndeed, with $t=\tau {\sigma }^2\in [0,2]$, the telescoping geometric series and $|1-t|\le 1$ give $|1-(1-t{)}^n|\le \min (nt,2)$. Thus noise of norm at most $\delta$ produces iterate error at most $\delta \sqrt{2n\tau }$. If $\tau \Vert A{\Vert }^2\le 1$, replace 2 by 1 to obtain $\delta \sqrt{n\tau }$. Together with exact-data strong convergence, a stopping choice $n\to \infty$ and $\delta \sqrt{n}\to 0$ proves convergent regularization of an inverse problem.

Two homotopy strips with the same endpoint paths and boundary on Lagrangian submanifolds glue to a relative $2$-cycle. If that cycle is zero in $H_2(M,L_0\cup L_1)$, its symplectic area vanishes: apply Stokes theorem, using $d\omega =0$ and the zero restriction of $\omega$ to each Lagrangian. Thus relative homology controls the ambiguity in a Lagrangian path-area functional.

There is $\mathcal{F}\subseteq \mathcal{P}({\omega }_1)$ of size ${\aleph }_2$ such that $\{X\cap \alpha :X\in \mathcal{F}\}$ is countable for every $\alpha <{\omega }_1$. Its initial-segment tree, pruned and leveled to be normal, is a Kurepa tree.

There exists an ${\omega }_1$-tree with countable levels and at least ${\aleph }_2$ distinct cofinal branches.

Let $\kappa$ be strongly inaccessible cardinal in the ground model and perform the finite Lévy collapse to omega-one. The ground full binary set-theoretic tree of height $\kappa$ has all levels of size less than $\kappa$, so these levels become countable. Its at least $({\kappa }^{+}{)}^M$ ground branches remain distinct. The chain condition for forcing preserves $({\kappa }^{+}{)}^M$, which is the extension ${\aleph }_2$. The unchanged ground set-theoretic tree therefore witnesses the Kurepa hypothesis.

For every natural parameter $k$, there is a finite length $N$ such that every sequence of $N$ finite rooted trees satisfying $|T_i|\le k+i$ has an earlier tree homeomorphically embedding into a later tree. It is a true finite-combinatorial principle unprovable in Peano arithmetic and in arithmetical transfinite recursion theory. Each fixed-parameter instance can still be verified by a sufficiently large finite search.

Finite rooted trees labelled in a well-quasi-ordering are a well-quasi-ordering under label-monotone tree embedding. A minimal bad sequence argument makes the collection of proper rooted subtrees a well-quasi-ordering; Higman lemma then compares their child lists while the root labels are compared in the label order.

If $I$ is an $\mathfrak{m}$-primary ideal generated by $n>0$ elements in a Noetherian local ring and $L=\mathrm{length}(R/I)$, then $I^j/I^{j+1}$ is a quotient of $\left(\genfrac{}{}{0ex}{}{j+n-1}{n-1}\right)$ copies of $R/I$. Summing their lengths gives the displayed bound. Since $I\subseteq \mathfrak{m}$, it also bounds $\mathrm{length}(R/{\mathfrak{m}}^t)$. Together with the prime-chain lower bound for local length, localization at a minimal prime proves the Krull height theorem.

The lengths of a finite uniquely decodable code satisfy ${\sum }_a{2}^{-l(a)}\le 1$, even when the code is not a prefix code. Distinct concatenations of $r$ codewords remain distinct; counting the possible binary strings of each total length bounds $({\sum }_a{2}^{-l(a)}{)}^r$ by $r{\max }_{a}l(a)$. Taking $r$th roots proves the claim.

For a bounded smooth solution of $u_t=Lu$ with initial function $f$, apply the Itô formula to $u(t-s,X_s)$. Its drift vanishes and its boundedness upgrades the localized stochastic integral to a true martingale. Taking endpoint expectations yields $u(t,x)={\mathbb{E}}_{x}f(X_t)$ without a global bound on $u_x$.

The functions $h_n(t)=\sqrt{2}\sin ((n-1/2)\pi t)$ give the Brownian covariance eigenbasis on $[0,1]$, with eigenvalues ${\lambda }_n=((n-1/2)\pi {)}^{-2}$. Integrating by parts against $g_n=\sqrt{2}\cos ((n-1/2)\pi t)$ gives independent coefficients ${\xi }_n=\int g_{n}dW$. Pathwise L2 completeness yields $W={\sum }_n\sqrt{{\lambda }_n}{\xi }_n{h}_n$ in L2 and the squared-norm series.

At an isolated interface with continuous base velocity, a stable density drop $\Delta \rho /2$ and a jump $[U^{\prime }]$ of base shear, decaying disturbances $w\propto e^{-k|z-z_i|}$ obey the displayed dispersion relation. It follows by inserting their derivative jump $[w^{\prime }]=-2kw$ into the jump conditions for stratified inviscid shear flow. Both the density jump and the vorticity jump contribute to the intrinsic phase velocity $c-U_i$.

For ${\psi }^{\prime \prime }={\alpha }^2\psi$ on $-1<z<1$ with endpoint values $P=\psi (1)$ and $Q=\psi (-1)$, the interior derivatives are ${\psi }^{\prime }(1)=\alpha [\coth (2\alpha )P-\mathrm{csch}(2\alpha )Q]$ and ${\psi }^{\prime }(-1)=\alpha [\mathrm{csch}(2\alpha )P-\coth (2\alpha )Q]$. Fitting the hyperbolic-function solution proves the map. It turns matching conditions into a finite-dimensional interface system, with exponentially small cross-interface coupling at large $\alpha$.