E.H. Shiguemori, H.F. de Campos Velho, J.D.S. Silva (2001): Artificial Neural Network in Heat Conduction Inverse Problem: Retrieval of Boundary Condition, Workshop of the Applied Computing Courses of the INPE ( WORKCAP-2001), 25 October, Sao Jose dos Campos (SP), Brasil, pp. 54-56.

Abstract: The goal of this work is the study of the use of two artificial neural networks, Multilayer Perceptron with back-propagation learning, and Radial Base Function, in the solution of an inverse problem recovering of boundary condition. Two data sets were used in the learning process, similar and non-similar functions. The generalization of trained networks with different data sets is compared. Another experiment tests the generalization with addition of noise.