A globally convergent primal-dual interior-point 3D filter method for nonlinear SDP

This paper proposes a primal-dual interior-point filter method for nonlinear semidefinite programming, which is the first multidimensional (three-dimensional) filter methods for interior-point methods, and of course for constrained optimization. A freshly new definition of filter entries is proposed, which is greatly different from those in all the current filter methods. A mixed norm is used to tackle with trust region constraints and global convergence to first-order critical points can easily be proved by slightly modifying the analysis in Ulbrich et al.\cite{Ulbrich-04}.