Global optimization: a naive approach

A unified approach is proposed for finding the global optimum of a continuously differentiable function phi :R/sup n/ to R of n real variables x=(x/sub 1/, x/sub 2/, . . ., x/sub n/) in R/sup n/, for both constrained and unconstrained problems. This approach can be implemented either numerically or in real time via an associated nonlinear programming circuit which is described. >