Robert Fung and Kuo-Chu Chang. Weighing and integrating evidence for stochastic simulation in Bayesian networks. In Proceedings of the Fifth Conference on Uncertainty in AI, pages 209--219, Windsor, 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....impractical [Pradhan et al. 1994; Shwe et al. 1991] This complexity result encouraged researchers to explore approximate inference, in particular Monte Carlo simulation and search techniques. The simulation methods include Gibbs sampling (straight sampling) Pearl, 1987] likelihood weighting [Fung Chang, 1990; Shachter Peot, 1990] logic sampling [Henrion, 1988] and randomized approximation schemes [Chavez Cooper, 1990] Many variations of these algorithms have been reported that improve on the run times [Fung Chang, 1990; Fung Del Favero, 1994; Hulme, 1995; Shachter Peot, 1990; Shwe ....

.... Gibbs sampling (straight sampling) Pearl, 1987] likelihood weighting [Fung Chang, 1990; Shachter Peot, 1990] logic sampling [Henrion, 1988] and randomized approximation schemes [Chavez Cooper, 1990] Many variations of these algorithms have been reported that improve on the run times [Fung Chang, 1990; Fung Del Favero, 1994; Hulme, 1995; Shachter Peot, 1990; Shwe Cooper, 1991] Dagum and Luby [Dagum Luby, 1993] showed that the general problem of approximate inference in belief networks with evidence is also NP hard. There are, however, restricted classes of networks in which ....

[Article contains additional citation context not shown here]

Fung, R. & Chang, K. C. (1990). Weighing and integrating evidence for stochastic simulation in Bayesian networks.

....and Spiegelhalter [30] Jensen [23] improved aspects of the algorithm which was proposed in [30] With approximate algorithms, based on a simulation of the corresponding Bayesian network. We mention the algorithms introduced by Chavez and Cooper [3] Dagum and Horvitz [6] Fung and Chang [12], Henrion [17] Hryceij [20] Jensen et al. 21] Pearl [36] Shachter and Peot [43] and Shwe and Cooper [44] 4 Our interest in obtaining an optimal decomposition originates in the evidence propagation algorithm proposed by Lauritzen and Spiegelhalter [30] This algorithm consists of the ....

R.M. Fung and K.C. Chang, Weighing and integrating evidence for stochastic simulation in Bayesian networks, in: M. Henrion, R.D. Shachter, L.N. Kanal and J.F. Lemmer, eds., Uncertainty in Articial Intelligence 5 (Elsevier, Amsterdam, 1990) 209-220.

....with the filtering algorithm above is that the samples generated at time t 1 are not influenced by observation o t 1 , which often allows particles to drift from the true belief state. Since we assume a DBN representation of dynamics, partial evidence integration (EI) or arc reversal [8] can be used to partially alleviate this problem [13] The generic structure of a DBN (assuming a fixed action) is shown in Figure 2(a) reversing the arc from S t 1 to O t 1 results in a network shown in Figure 2(b) With this structure, given a particle s t (i) and observation o t 1 , a ....

Robert Fung and Kuo-Chu Chang. Weighing and integrating evidence for stochastic simulation in Bayesian networks. In Proceedings of the Fifth Conference on Uncertainty in AI, pages 209--219, Windsor, 1989.

....These algorithms comprise simulation based and searchbased approximations. Simulation based algorithms use a source of random bits to generate random samples of the solution space. Simulation based algorithms include straight simulation [25, 26] forward simulation, 15] likelihood weighting [13, 33], and randomized approximation schemes [3, 4, 7, 8] Variants of these methods such as backward simulation [14] exist; Neal [23] provides a good overview of the theory of simulation based algorithms. Search based algorithms search the space of alternative instantiations to find the most probable ....

....times. We construct the bounded variance algorithm that proves that the complexity of approximating inferences in belief networks without extreme conditional probabilities is polynomial time solvable. The bounded variance algorithm is a simple variant of the known likelihoodweighting algorithm [13, 33], which employs recent results on the design of optimal algorithms for Monte Carlo simulation [9] We consider an n node belief network without extreme conditional probabilities and an evidence set E of constant size. We prove that, with a small failure probability ffi, the bounded variance ....

R. Fung and K.-C. Chang. Weighing and integrating evidence for stochastic simulation in Bayesian networks. In M. Henrion, R. Shachter, L. Kanal, and J. Lemmer, editors, Uncertainty in Artificial Intelligence 5, pages 209--219. Elsevier, Amsterdam, The Netherlands, 1990. 38

....are examinations of resource bounded algorithms for belief networks. Horvitz et al. 89] employs bounded conditioning, a technique we believe may perform well in the OLMA and which we hope to include in some future investigations. We likewise will seek competitive forms of stochastic simulation [Fung Chang, 89] and continue our explorations with the kappa calculus [Goldszmidt, 95] Conclusions We are interested in developing and characterizing decision algorithms with robust real time performance. We presented the On Line Maintenance domain, a domain we think is uniquely suited for effective ....

R. Fung, and K. Chang. Weighing and Integrating Evidence for Stochastic Simulation in Bayesian Networks. In Proceedings of the Fifth Conference on Uncertainty in AI. pp 112-117, August, 1989.

....Jensen (1994) improved aspects of the algorithm which was proposed in Lauritzen and Spiegelhalter (1988) ffl With approximate algorithms, based on a simulation of the corresponding Bayesian network. We mention the algorithms introduced by Chavez and Cooper (1990) Dagum and Horvitz (1993) Fung and Chang (1990), Henrion (1988) Hryceij (1990) Jensen et al. 1993) Pearl (1987) Shachter and Peot (1990) and Shwe and Cooper (1991) We take as point of departure the evidence propagation algorithm proposed by Lauritzen and Spiegelhalter (1988) The first step of this algorithm consist of the moralization ....

Fung, R.M. and Chang, K.C. (1990) Weighing and integrating evidence for stochastic simulation in Bayesian networks, in Uncertainty in Artificial Intelligence 5, Henrion, M., Shachter, R.D., Kanal, L.N. and Lemmer, J.F. (eds), Elsevier, Amsterdam, pp. 209-220.

....probabilistic networks (DPNs) 5, 11, 10] DPNs form an important class of BNs for modeling dynamical systems and sequential decision processes. Because of their size, exact methods are often rejected in favor of simulation techniques. In the case of DPNs, arc reversal or evidence integration [7] is extremely important; this case has been made forcefully [10] However, even partial evidence integration can cause a large blowup in the size of CPTs; hence structured arc reversal can play an important role. We also show how the reversed DPNs can exploit the structured CPTs in simulation ....

....a BN so that the arc between two nodes has its directionality reversed, while still correctly representing the original distribution. Arc reversal is an important technique for BNs and influence diagrams, and plays a significant role in the evaluation of BNs through stochastic simulation [7, 10], as we describe in the next section. The basic arc reversal operation is relatively straightforward. Consider a network where variable A is a parent of O. The variables belonging to the set Pi(A) Pi(O) can be divided into three classes: X = Pi(A) n Pi(O) Y = Pi(A) Pi(O) and Z = ....

[Article contains additional citation context not shown here]

R. Fung and K. Chang. Weighing and integrating evidence for stochastic simulation in bayesian networks. UAI-89, pp.209--219, Windsor.

....the time complexity of the known algorithms increases with the degree of connectivity of the network. For large multiply connected networks approximation algorithms are often used, either based on stochastic simulation, e.g. Chavez and Cooper, 1990; Chavez, 1990; Dagum and Chavez, 1991; Fung and Chang, 1990; Henrion, 1987; Pearl, 1987; Shachter and Peot, 1990) or search through the space of alternative instantiations, e.g. Cooper, 1984; Henrion, 1990; Henrion, 1991; Peng and Reggia, 1987a; Peng and Reggia, 1987b) In practice these algorithms allow one to reason with more complex networks than ....

Fung, R. and Chang, K. (1990). Weighing and integrating evidence for stochastic simulation in Bayesian networks. In (Henrion et al., 1990), pages 209--219.

....developed by Henrion [8] and is called logic sampling. It works well, when there is no given evidence on the graph, but the number of necessary runs to achieve a give precision increases exponentially with the number of observations. Likelihood weighting algorithms were proposed by Fung and Chang [7] and Shachter and Peot [21] In general these algorithms have a good performance, but one can determine very simple examples in which the complexity is the same than in Henrion s algorithm: Consider that we have evidence in one variable which has an only parent and that the conditional ....

Fung, R., and Chang, K.C., Weighing and integrating evidence for stochastic simulation in bayesian networks. In: Uncertainty in Artificial Intelligence, 5 (M. Henrion, R.D. Shacter, L.N. Kanal, J.F. Lemmer, eds.) North-Holland, Amsterdam, 209--220, 1990.

No context found.

Robert Fung and Kuo-Chu Chang. Weighing and integrating evidence for stochastic simulation in Bayesian networks. In Proceedings of the Fifth Conference on Uncertainty in AI, pages 209--219, Windsor, 1989.

No context found.

Robert Fung and Kuo-Chu Chang. Weighing and integrating evidence for stochastic simulation in Bayesian networks. In Proceedings of the Fifth Conference on Uncertainty in AI, pages 209--219, Windsor, 1989.

No context found.

R. Fung and K Chang. Weighing and integrating evidence for stochastic simulation in bayesian networks. In Proc. 5th Conf. Uncertainty in Arti#cal Intelligence, pages 209#219, 1989.

No context found.

R. Fung and K.-C. Chang. Weighing and integrating evidence for stochastic simulation in Bayesian networks. In M. Henrion, R. Shachter, L. Kanal, and J. Lemmer, editors, Uncertainty in Artificial Intelligence 5, pages 209--219, New York, N. Y., 1989. Elsevier Science Publishing Company, Inc.

No context found.

Robert Fung and Kuo-Chu Chang. Weighing and integrating evidence for stochastic simulation in Bayesian networks. In Uncertainty in Arti cial Intelligence 5, pages 209-219, New York, N. Y., 1989. Elsevier Science Publishing Company, Inc.

No context found.

Robert Fung and Kuo-Chu Chang. Weighing and integrating evidence for stochastic simulation in Bayesian networks. In Uncertainty in Artificial Intelligence 5, pages 209-219, New York, N. Y., 1989. Elsevier Science Publishing Company, Inc., 1989.

No context found.

Fung, R. and K. C. Chang. Weighing and integrating evidence for stochastic simulation in bayesian networks, In: Uncertainty in Artificial Intelligence 5, North-Holland, Amsterdam, 1990.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC