This paper presents a model of neural network embodiment of intentions and planning mechanisms for autonomous agents. The model bridges the dichotomy of symbolic and non-symbolic ...
— We have found a more general formulation of the REINFORCE learning principle which had been proposed by R. J. Williams for the case of artificial neural networks with stochast...
In this paper a new approach for approximation problems involving only few input and output parameters is presented and compared to traditional Backpropagation Neural Networks (BP...
Paper [1] aimed at providing a unified presentation of neural network architectures. We show in the present comment (i) that the canonical form of recurrent neural networks presen...
This research introduces a general class of functions serving as generalized neuron models to be used in artificial neural networks. They are cast in the common framework of comp...