Y. El Fattah, R. Dechter, Diagnosing tree-decomposable circuits, in: Proc. IJCAI-95, Montreal, Quebec, 1995, pp. 1742--1748.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....process is a single pass algorithm from the leaves of the tree to the root, whereas Stumptner and Wotawa need a two pass approach since they must first enumerate all component diagnoses. A second major tree based method uses a clique tree decomposition of a system, e.g. the diagnostic method of [13].A clique tree is a representation that is used for many kinds of inference in addition to diagnosis, including probabilistic inference and constraint satisfaction. The tree we generate is a directed tree with a fixed root, and the nodes of the tree are generated based on the clan property; a ....

Yousri El Fattah and Rina Dechter. Diagnosing treedecomposable circuits. In IJCAI, pages 1742--1749, 1995.

....choice was made. Constraint based network management tools do address this problem. Based on a model based approach [Hamscher, Console, de Kleer, 1992] constraint technology has been used to diagnose communication protocols [Riese, 1993] and special classes of randomly generated circuits [El Fattah Dechter, 1995] . More recently, constraints have been applied to routing to solve bandwidth allocation planning [Frei Faltings, 1999] and QoS based multi domain routing [Calisti Faltings, 1999] In our work, we have extended the problem domain to encompass fault management of Internet network services ....

El Fattah, Y., and Dechter, R. 1995. Diagnosing tree-decomposable circuits. In Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, 1742--1748.

....[Cadoli and Donini, 1997] Sometimes we will be interested in theories which are logically equivalent to P I LT ( but which need not include all L T implicates, and can thus be much more concise. We refer to any such theory as a L T LUB of , following [Selman and Kautz, 1996] see also [del Val, 1995] . We ll see one particular application of L T LUBs in our discussion of diagnosis later. 2 Kernel resolution: Review Kernel resolution, described in [del Val, 1999; 2000a] is a consequence finding generalization of ordered resolution. We assume a total order of the propositional variables x ....

.... others: chandra n [Chandra and Markowsky, 1978] and mxky [Kean and Tsiknis, 1990] are problems with known exponential behavior; adder n (n bit adders) serial adder (n bit adders obtained by connecting n full adders) bnp n p (tree structured circuits with n layers, where p is a probability, [El Fattah and Dechter, 1995]) represent digital circuits; serial diag adder and bnp ab represent the same circuits in a form suitable for diagnostic reasoning; pipes and pipes syn are problems drawn from qualitative physics; type k n problems, 1 On the Dimacs [Dimacs, 1992] machine scale benchmark, this Pentium II 400MHz ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In IJCAI'95, Proc. 14th Int. Joint Conf. on Artificial Intelligence, pp. 1742--1748, 1995.

....starting the diagnosis and measurement selection process. The diagnosis kernel, i.e. the class library for the user interface, the diagnosis engine, and the compiler, was implemented in Visualworks Smalltalk. Diagnosis and measurement selection runtimes are competitive with other implementations [9, 21, 5]. Parts of our VHDL debugger [8, 20, 22] were implemented using the diagnosis kernel. 4 The AD2L Language The purpose of designing a dedicated system description language for model based diagnosis is to support the user in writing the actual models. He should not be required to engage in ....

Yousri El Fattah and Rina Dechter, `Diagnosing tree-decomposable circuits ', in Proceedings 14 th International Joint Conf. on Artificial Intelligence, pp. 1742 -- 1748, (1995).

....efficient constraint propagation, and model compilation, each individually yielded at least a 10 fold improvement in diagnostic efficiency over a simple, one level model of structure and behaviour. Another example of the improved model efficiency is the use of structure based abduction [El Fattah and Dechter, 1994]. First, a model has to be transformed into an acyclic constraint network. This might require a relatively expensive pre processing of the initial constraints using tree clustering. However, as a consequence, an equivalent of the Min diag procedure is replaced by a single model call which returns ....

El Fattah, Y., Dechter, R. Diagnosing tree-decomposable circuits. Technical Report 94-18, Univ. of California at Irvine, CA, 1994.

....strategies (e.g. SD89, GRST89] However, fault probabilities are not available in all domains, and complete discrimination between diagnoses is also not possible in all examples. Therefore in the past years, much effort has been spent on the development of faster diagnosis algorithms. In [FD95], an algorithm based on constraint satisfaction for tree structured diagnosis systems was introduced whose performance, as the accompanying experimental evaluation showed, compared favorably with the algorithms of Reiter [Rei87, GSW89] and de Kleer [dKW87] A new algorithm for the diagnosis of ....

....figures are based on searching for single faults. There are no further restrictions on the internal structure of the circuit (i.e. no symmetry or constant depth of leaves) To allow a direct comparison of TREE DIAG with other existing MBD algorithms, Reiter s HS DAG and the SAB algorithm from [FD95] were also implemented. Figure 3 shows the amount of diagnosis time as a function of the number of components for circuits with the number of inputs set at half the number of internal components. Results were derived by averaging over 100 different example runs. Comparing the results leads to the ....

[Article contains additional citation context not shown here]

Yousri El Fattah and Rina Dechter. Diagnosing tree-decomposable circuits. In Proceedings 14 th International Joint Conf. on Artificial Intelligence, pages 1742 -- 1748, 1995.

No context found.

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742-- 1748, Montreal, Canada, August 1995.

....error ratios M=L and U=M ) The efficiency gain is represented by the time ratio TR = T e =T a , where T e is the running time for elim mpe and T a is the running time for approx mpe. We also report the width, w, and the induced width, w , of the network s graph along the min degree 3 ordering [3, 25] used by the algorithms. Remember that for i w , approx mpe(i) coincides with elim mpe. For diagnostic purposes we also report the maximum number of mini buckets created in a single bucket, max mb (we report averages where max mb is rounded to the nearest integer) We present empirical results ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742--1748, Montreal, Canada, August 1995.

....requirement. Algorithm tree clustering and cycle cutset conditioning are two extremes in this spectrum. We investigate the potential of our scheme in the domain of combinatorial circuits. This domain is frequently used as an application area in both probabilistic and deterministic reasoning [18, 40, 16]. We analyze 11 benchmark combinatorial circuits widely used in the fault diagnosis and testing community [6] see Table 1 ahead. For each circuit, the analysis is summarized in a chart displaying the time space complexity tradeo#s for diagnosing that circuit. The analysis allows tailoring the ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742--1748, Montreal, Canada, August 1995.

....and T a was the running time for approx mpe. When elim mpe was infeasible, we report only the ratio U=L. Note, that U=L is an upper bound on the error ratios M=L and U=M . We also recorded the width, w, and the induced width, w 3 , of the network s graph, along the min degree 4 ordering [3, 20] used by the algorithms. Remember, that for i w 3 approx mpe(i) coincides with elim mpe. For diagnostic purposes, we also report the maximum number of mini buckets recorded, max mb. We observe that in many cases the mini bucket scheme quickly finds a relatively close approximation to the ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742--1748, Montreal, Canada, August 1995.

....and T a was the running time for approx mpe. When elim mpe was infeasible, we report only the ratio U=L. Note, that U=L is an upper bound on the error ratios M=L and U=M . We also recorded the width, w, and the induced width, w 3 , of the network s graph, along the mindegree 4 ordering [3, 20] used by the algorithms. Remember, that for i w 3 approx mpe(i) coincides with elim mpe. For diagnostic purposes, we also report the maximum number of mini buckets recorded, max mb. We observe that in many cases the mini bucket scheme quickly finds a relatively close approximation to the ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742--1748, Montreal, Canada, August 1995.

....given number of time slices determined by the up and down time constraints. These clique chain structures are common in many temporal domains that possess the Markov property (the future is independent of the past given the present) Another example of structured domain is circuit diagnosis. In [27] it was shown that circuit diagnosis benchmarks can be embedded in a tree of cliques, where the clique sizes are substantially smaller than the overall number of variables. In general, one can imagine a variety of real life domains having such structure that is captured by k tree embeddings [1] ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742--1748, Montreal, Canada, August 1995.

....and T a was the running time for approx mpe. When elim mpe was infeasible, we report only the ratio U=L. Note, that U=L is an upper bound on the error ratios M=L and U=M . We also recorded the width, w, and the induced width, w 3 , of the network s graph, along the mindegree 4 ordering [3, 20] used by the algorithms. Remember, that for i w 3 approx mpe(i) coincides with elim mpe. For diagnostic purposes, we also report the maximum number of mini buckets recorded, max mb. We observe that in many cases the mini bucket scheme quickly finds a relatively close approximation to the ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742--1748, Montreal, Canada, August 1995.

....and T a was the running time for approx mpe. When elim mpe was infeasible we report only the ratio U=L. 28 Note, that U=L is an upper bound on the error ratios M=L and U=M . We also recorded the width, w, and the induced width, w , of the network s graph along the min degree 3 ordering [3, 21] used by the algorithms. Remember that for i w approx mpe(i) coincides with elim mpe. For diagnostic purposes we also report the maximum number of mini buckets recorded, max mb. We observe that in many cases the mini bucket scheme quickly finds a relatively close approximation to the MPE ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742--1748, Montreal, Canada, August 1995.

....requirement. Algorithm tree clustering and cycle cutset conditioning are two extremes in this spectrum. We investigate the potential of our scheme in the domain of combinatorial circuits. This domain is frequently used as an application area in both probabilistic and deterministic reasoning [6 8]. We analyze 11 benchmark combinatorial circuits widely used in the fault diagnosis and testing community [9] see Table 1 ahead. For each circuit, the analysis is summarized in a chart displaying the time space complexity tradeoffs for diagnosing that circuit. The analysis allows tailoring the ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In International Joint Conference of Artificial Intelligence (IJCAI-95), pages 1742--1748, Montreal, Canada, August 1995.

....their hybrids) to real life applications. To that end we investigate empirically their potential in the domain of processing combinatorial circuits. This domain is frequently used as an application area in both probabilistic and deterministic reasoning [ Geffner and Pearl, 1987; Srinivas, 1994; El Fattah and Dechter, 1995 ] The experiments are conducted on 11 benchmark combinatorial circuits widely used in the fault diagnosis and testing community [ Brglez and Fujiwara, 1985 ] See Table 1. These experiments allow us to assess in advance by graph manipulation only the complexity of diagnosis and abduction tasks ....

Y. El Fattah and R. Dechter. Diagnosing tree-decomposable circuits. In Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence (IJCAI95) , Montreal, Quebec, Canada, 1995.

No context found.

Y. El Fattah, R. Dechter, Diagnosing tree-decomposable circuits, in: Proc. IJCAI-95, Montreal, Quebec, 1995, pp. 1742--1748.

No context found.

Yousri El Fattah and Rina Dechter, `Diagnosing tree-decomposable circuits ', in Proceedings of IJCAI-95, pp. 1742--1749, (1995).

No context found.

Yousri El Fattah and Rina Dechter, `Diagnosing tree-decomposable circuits ', in Proceedings of IJCAI-95, pp. 1742--1749, (1995).

No context found.

Yousri El Fattah and Rina Dechter, `Diagnosing tree-decomposable circuits ', in Proceedings of IJCAI-95, pp. 1742--1749, (1995).

No context found.

Yousri El Fattah and Rina Dechter, `Diagnosing tree-decomposable circuits ', in Proceedings of IJCAI-95, pp. 1742--1749, (1995).

No context found.

Yousri El Fattah and Rina Dechter, `Diagnosing tree-decomposable circuits ', in Proceedings of IJCAI-95, pp. 1742--1749, (1995).

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