zbMATH — the first resource for mathematics

Learning pseudo-independent models: Analytical and experimental results. (English) Zbl 0987.68529
Hamilton, Howard J. (ed.), Advances in artificial intelligence. 13th biennial conference of the Canadian Society for Computational Studies of Intelligence, AI 2000, Montréal, Quebec, Canada, May 14-17, 2000. Proceedings. Berlin: Springer. Lect. Notes Comput. Sci. 1822, 227-239 (2000).
Summary: Most algorithms to learn belief networks use single-link lookahead search to be efficient. It has been shown that such search procedures are problematic when applied to learning pseudo-independent (PI) models. Furthermore, some researchers have questioned whether Pl models exist in practice. We present two non-trivial PI models which derive from a social study dataset. For one of them, the learned PI model reached ultimate prediction accuracy achievable given the data only, while using slightly more inference time than the learned non-PI model. These models provide evidence that PI models are not simply mathematical constructs. To develop efficient algorithms to learn PI models effectively we benefit from studying and understanding such models in depth. We further analyze how multiple PI submodels may interact in a larger domain model. Using this result, we show that the RML algorithm for learning PI models can learn more complex PI models than previously known.
For the entire collection see [Zbl 0939.00040].

68T05 Learning and adaptive systems in artificial intelligence
68P20 Information storage and retrieval of data
68T37 Reasoning under uncertainty in the context of artificial intelligence