This paper deals with the determination of the rate of convergence to the u
nit of some multivariate neural network operators, namely, the Cardaliaguet
-Euvrard and "squashing" operators. This is given through the multidimensio
nal modulus of continuity of the involved multivariate function or its part
ial derivatives of specific order that appear in the right-hand side of the
associated multivariate jackson type inequality. (C) 2000 Elsevier Science
Ltd. All rights reserved.