Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $v_1,v_2,\dots,v_n,u_1,u_2,\dots,u_r\in\mathbb{R}^n$. Can one find analytical conditions (not write the problem up as a convex optomisation problem and argue it can be solved this way) under which there exists $\alpha_1,\alpha_2,\dots,\alpha_r>0$ such that

$\sum_{i=1}^r\alpha_iu_i\in K=\{\sum_{j=1}^n\theta_jv_j:\theta_j>0\}$.

That is, conditions such that there exists some (strictly?) conical combination (sorry, not really sure what the term for this is) of vectors $u_1,\dots,u_r$ that lies in the inside of the cone $K$ defined by $v_1,\dots,v_n$. Clearly, sufficient conditions are that for some $i$, $u_i\in K$, I'm looking for something a bit more insightful (ideally necessary and sufficient).

Thanks.

share|improve this question

1 Answer 1

up vote 0 down vote accepted

A first idea: It's about the intersection of two open cones, the first being $K$ and the second $C:=\{\sum_{i=1}^r \alpha_i u_i\ |\ \alpha_i>0\ (1\leq i\leq r)\}$. When they don't intersect (apart from $0$) there is a separating hyperplane, i.e., a nonzero vector $a$ such that $a\cdot u_i\geq 0$ for all $i$ and $a\cdot v_j\leq 0$ for all $j$.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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