This paper introduces a new technique for the alleviation of local minima i
n minimization problems. The proposed "stretching" technique transforms the
objective function by stretching upwards the neighborhood of a given point
, and assists in eliminating local minima, while preserving the global ones
. Experiments indicate that when a global search method converges to an und
esired local minimum, the use of the new technique provides a way of escape
and helps convergence to the global minimum.