Given a d-polytope $P$ we define the c-dual polytope as $P^\ast = \{y\in R^d \mid x\cdot y\geq -c, \forall x\in P\}$. Then I say that a polytope is c-polar self-dual if $P=P^\ast$. I cannot find this definition of self-dual used in any research level publications and before I publish I would like to know who I need to cite. The only other place I have seen this discussed is in this question:

Does there always exist a self dual polytope that contains a given polytope contained in its dual?

  • 1
    $\begingroup$ Judging by Google search results, self-dual cones were studied quite a bit. Since self-dual bodies in $R^d$ are precisely compact hyperplane sections of self-dual cones in $R^{d+1}$, it should be possible to translate many results from one context to the other. $\endgroup$ Commented Dec 5, 2013 at 20:23


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.