A new connectionist model for the solution of piecewise lin- ear regression problems is introduced; it is able to reconstruct both con- tinuous and non continuous real valued mappings starting from a finite set of possibly noisy samples. The approximating function can assume a different linear behavior in each region of an unknown polyhedral parti- tion of the input domain. The proposed learning technique combines local estimation, clustering in weight space, multicategory classification and linear regression in order to achieve the desired result. Through this approach piecewise affine solutions for general nonlinear regression problems can also be found.