Constrained Search in Imaginary Time

Abstract: An optimization method for the expectation value of a self-adjoint operator under a finite number of expectation-value constraints based on imaginary-time evolution is introduced. It is formulated for finite-dimensional Hilbert spaces and uses linearly independent and commuting self-adjoint operators for the constraints. The method is applied to the problem of finding the universal functional of density-functional theory and allows theoretical insights into the density-potential mapping.