Adaptive Probabilistic Networks

Résumé du livre

Abstract: "Belief networks and neural networks are two forms of network representations that have been used in the development of intelligent systems. Belief networks utilize the causal structure inherent in most domains to allow concise representation of a probability distribution reflecting the uncertainty about the world. They facilitate exact calculation of the probability of propositions of interest. Neural networks combine simple neuron-like units to represent complex functions. The simple, local nature of most neural network training algorithms provides a certain biological plausibility and allows for a massively parallel implementation. In this paper, we show that similar local learning algorithms can be derived for belief networks, and that these learning algorithms can operate using only information that is directly available from the normal, inferential processes of the networks. This removes the main obstacle preventing belief networks from competing with neural networks on the above-mentioned tasks. The precise, local, probabilistic interpretation of belief networks also allows them to be partially or wholly constructed by humans; allows the results of learning to be easily understood; and allows them to contribute to rational decision-making in a well-defined way."