Hill-climbing and branch-and-bound algorithms for exact and approximate inference in credal networks

被引:19
|
作者
Cano, Andres [1 ]
Gomez, Manuel [1 ]
Moral, Serafin [1 ]
Abellan, Joaquin [1 ]
机构
[1] Univ Granada, ETS Ingn Informat, Dept Comp Sci & Artificial Intelligence, E-18071 Granada, Spain
关键词
credal network; probability intervals; Bayesian networks; strong independence; hill-climbing; branch-and-bound algorithms;
D O I
10.1016/j.ijar.2006.07.020
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper proposes two new algorithms for inference in credal networks. These algorithms enable probability intervals to be obtained for the states of a given query variable. The first algorithm is approximate and uses the hill-climbing technique in the Shenoy-Shafer architecture to propagate in join trees; the second is exact and is a modification of Rocha and Cozman's branch-and-bound algorithm, but applied to general directed acyclic graphs, (C) 2006 Elsevier Inc. All rights reserved.
引用
收藏
页码:261 / 280
页数:20
相关论文
共 50 条
  • [1] Application of a hill-climbing algorithm to exact and approximate inference in credal networks
    Cano, Andres
    Gomez, Manuel
    Moral, Serafin
    [J]. ISIPTA 05-PROCEEDINGS OF THE FOURTH INTERNATIONAL SYMPOSIUM ON IMPRECISE PROBABILITIES AND THEIR APPLICATIONS, 2005, : 88 - 97
  • [2] Inference in credal networks: branch-and-bound methods and the A/R plus algorithm
    da Rocha, JCF
    Coman, FG
    [J]. INTERNATIONAL JOURNAL OF APPROXIMATE REASONING, 2005, 39 (2-3) : 279 - 296
  • [3] Approximate Inference in Logical Credal Networks
    Marinescu, Radu
    Qian, Haifeng
    Gray, Alexander
    Bhattacharjya, Debarun
    Barahona, Francisco
    Gao, Tian
    Riegel, Ryan
    [J]. PROCEEDINGS OF THE THIRTY-SECOND INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2023, 2023, : 5694 - 5701
  • [4] A hill-climbing learning method for Hopfield networks
    Tang, Z
    Jin, HH
    Murao, K
    Ishizuka, O
    Tanno, K
    [J]. ELECTRONICS AND COMMUNICATIONS IN JAPAN PART III-FUNDAMENTAL ELECTRONIC SCIENCE, 2001, 84 (07): : 28 - 40
  • [5] Finite Exact Branch-and-Bound Algorithms for Concave Minimization over Polytopes
    Marco Locatelli
    Nguyen V. Thoai
    [J]. Journal of Global Optimization, 2000, 18 : 107 - 128
  • [6] Finite exact branch-and-bound algorithms for concave minimization over polytopes
    Locatelli, M
    Thoai, NV
    [J]. JOURNAL OF GLOBAL OPTIMIZATION, 2000, 18 (02) : 107 - 128
  • [7] Genetic and hill-climbing algorithms for laser beam correction
    Sheldakova, J
    Rukosuev, A
    Kudryashov, A
    [J]. LASER RESONATORS AND BEAM CONTROL VII, 2004, 5333 : 106 - 111
  • [8] Approximate and branch-and-bound algorithms for the parallel machine scheduling problem with a single server
    Liu, Guo-Sheng
    Li, Jin-Jin
    Yang, Hai-Dong
    Huang, George Q.
    [J]. JOURNAL OF THE OPERATIONAL RESEARCH SOCIETY, 2019, 70 (09) : 1554 - 1570
  • [9] EXPECTED PERFORMANCE OF BRANCH-AND-BOUND ALGORITHMS
    LENSTRA, JK
    RINNOOYKAN, AHG
    [J]. OPERATIONS RESEARCH, 1978, 26 (02) : 347 - 349
  • [10] Real-coded memetic algorithms with crossover hill-climbing
    Lozano, M
    Herrera, F
    Krasnogor, N
    Molina, D
    [J]. EVOLUTIONARY COMPUTATION, 2004, 12 (03) : 273 - 302