This paper is concerned with the automatic induction of parsimonious neural networks. In contrast to other program induction situations, network induction entails parametric learning as well as structural adaptation. We present a novel representation scheme, called neural trees, that allows eecient learning of network architectures and parameters both by genetic search. A hybrid evolutionary me...

This paper is concerned with the automatic induction of parsimonious neural networks. In contrast to other program induction situations, network induction entails parametric learning as well as structural adaptation. We present a novel representation scheme called neural trees that allows efficient learning of both network architectures and parameters by genetic search. A hybrid evolutionary me...

The question of how the vertebrate embryo gives rise to a nervous system is of paramount interest in developmental biology. Neural induction constitutes the earliest step in this process and is tightly connected with development of the embryonic body axes. In the Xenopus embryo, perpendicular gradients of BMP and Wnt signals pattern the dorsoventral and anteroposterior body axes. Both pathways ...