Conditioning algorithms for exact and approximate inference in causal networks
作者： Adnan Darwiche
DOI:
关键词:
摘要: We present two algorithms for exact and approximate inference in causal networks. The first algorithm, dynamic conditioning, is a refinement of cutset conditioning that has linear complexity on some networks which exponential. second B-conditioning, an algorithm allows one to trade-off the quality approximations with computation time. also experimental results illustrating properties proposed algorithms.
参考文章(7)
Moises Goldszmidt, Fast belief update using order-of-magnitude probabilities uncertainty in artificial intelligence. pp. 208- 216 ,(1995)
Adnan Darwiche, Moisés Goldszmidt, Action networks: a framework for reasoning about actions and change under uncertainty uncertainty in artificial intelligence. pp. 136- 144 ,(1994) , 10.1016/B978-1-55860-332-5.50023-7
F.J. Díez, Local conditioning in Bayesian networks Artificial Intelligence. ,vol. 87, pp. 1- 20 ,(1996) , 10.1016/0004-3702(95)00118-2
H.Jacques Suermondt, Gregory F. Cooper, Probabilistic inference in multiply connected belief networks using loop cutsets International Journal of Approximate Reasoning. ,vol. 4, pp. 283- 306 ,(1990) , 10.1016/0888-613X(90)90003-K
Mark A Peot, Ross D Shachter, None, Fusion and propagation with multiple observations in belief networks Artificial Intelligence. ,vol. 48, pp. 299- 318 ,(1991) , 10.1016/0004-3702(91)90030-N
M. Goldszmidt, A. Darwiche, Plan simulation using Bayesian networks conference on artificial intelligence for applications. pp. 155- 161 ,(1995) , 10.1109/CAIA.1995.378777
Judea Pearl, Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference ,(1988)