TheoremBase

Lagrange Multipliers

At a local extremum of a C¹ function restricted to the zero set of m C¹ constraints, the gradient of the function is a linear combination of the gradients of the constraints, provided some m columns of the constraints' Jacobian matrix form an invertible matrix; for one constraint, provided its gradient does not vanish.

Statement

In the setting of Functions of Class Cᵏ: Notation, let f:U→Rf:U\to\mathbb{R} be of class C1C^{1} on UU, let M⊆UM\subseteq U, and let a∈Ma\in M be such that the restriction f∣M:M→Rf|_{M}:M\to\mathbb{R} has a local maximum or a local minimum at aa relative to MM, in the metric space (Rn,dE)(\mathbb{R}^{n},d_{E}).

Several constraints. Let m<nm<n, and let g:U→Rmg:U\to\mathbb{R}^{m} be of class C1C^{1} on UU with M={x∈U:g(x)=0}M=\{x\in U:g(x)=0\}. Suppose that some mm columns of the Jacobian matrix Dg(a)Dg(a) form an invertible matrix: there is an injective function σ:[m]→[n]\sigma:[m]\to[n] such that the matrix in Mm(R)\mathcal{M}_{m}(\mathbb{R}) with entries (Dg(a))k σ(l)\bigl(Dg(a)\bigr)_{k\,\sigma(l)} (k,l∈[m]k,l\in[m]) is invertible. Then there is λ∈Rm\lambda\in\mathbb{R}^{m} with

∇f(a)=Dg(a)⊤λ=∑k∈[m]λk ∇gk(a).\nabla f(a)=Dg(a)^{\top}\lambda=\sum_{k\in[m]}\lambda_{k}\,\nabla g_{k}(a).

One constraint. Let n≥2n\ge2, and let φ:U→R\varphi:U\to\mathbb{R} be of class C1C^{1} on UU with M={x∈U:φ(x)=0}M=\{x\in U:\varphi(x)=0\} and ∇φ(a)≠0\nabla\varphi(a)\neq0. Then there is λ∈R\lambda\in\mathbb{R} with

∇f(a)=λ ∇φ(a).\nabla f(a)=\lambda\,\nabla\varphi(a).

Proofs

Log in to submit a proof.

Loading...

Citations

Loading…

Dependencies

Loading…

Related

0 relations

Curated associations between results. These are editable and subjective — they do not replace the dependency graph, which is derived from the references in the text.

No relations recorded yet.

Comments

Log in to comment.

Loading…