Lecture Notes

14. Ellipsoids

Part of the Series on Linear Algebra.

By Akshay Agrawal. Last updated Dec. 20, 2018.

Previous entry: Matrix norms ; Next entry: Singular Value Decomposition

An ellipsoid centered at is a set

where is a self-adjoint, positive definite matrix. The lengths of the semi-axies of are given by , where the are the eigenvalues of To see why is an ellipsoid when , write , where the columns of are orthonormal eigenvectors of , and notice that

where . In particular, if , then

An alternate representation of an ellipsoid is

If is symmetric and positive definite, then taking gives the ellipsoid defined in the first representation. To see this for the case when , note that taking gives , which in turn equals .

References

  1. Stephen Boyd and Lieven Vandenberghe. Convex Optimization.