Let $F:\mathbb{R}^{n}\rightarrow \mathbb{R}$ be defined as \begin{align} F(\textbf{v})&=||\textbf{v}||\\ &=\sqrt{\sum\limits_{i=1}^{n}v_{i}^{2}}, ~\textbf{v}=[v_{1},\ldots,v_{n}]^{T}\in \mathbb{R}^{n}. \end{align} Let $L$ and $L'$ be two subsets of $\mathbb{R}^{n}$, with $L'\subset L$, defined as follows: \begin{align} L&=\{\textbf{x}\in \mathbb{R}^{n}:A\textbf{x}=\textbf{b}\},\\ L'&=\{\textbf{x}\in \mathbb{R}^{n}:A'\textbf{x}=\textbf{b}'\}; \end{align} here, $A$ is an $m\times n$ matrix, $\textbf{b}\in \mathbb{R}^{m}$, $A'$ is a $k\times n$ matrix where $k>m$ with the property that the first $m$ rows of $A'$ are those of $A$, and the first $m$ entries of $\textbf{b}'\in \mathbb{R}^{k}$ are those of $b$ (thus making $L'\subset L$). The matrices $A$ and $A'$ are of full rank.

Now, suppose that $\textbf{v}$ is the unique minimizer of $F(\textbf{x})$ subject to $\textbf{x}\in L$, i.e., $\textbf{v}$ is the unique point in $L$ where $F$ attains a minimum in $L$. Suppose further that $\textbf{v}\in L'$. I want to show that $\textbf{v}$ is the unique minimizer of $F(\textbf{x})$ subject to $\textbf{x}\in L'$.

I tried the method of Lagrange multipliers. I wrote down the necessary conditions that arise from Lagrange's theorem (given that we know $\textbf{v}$ is a minimizer of $F$ in $L$), but they are not taking me anywhere.

I'd appreciate any help or hints. Thanks.


Here's a hint: You have $L' \subset L$, so can there be an $\mathbf{x} \in L'$ with $F(\mathbf{x}) < F(\mathbf{v})$?

  • $\begingroup$ Great! Thanks! It was this simple! $\endgroup$ – PN Karthik Apr 6 '17 at 7:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.