Filters








143,176 Hits in 4.4 sec

Learning Chordal Markov Networks via Branch and Bound

Kari Rantanen, Antti Hyttinen, Matti Järvisalo
2017 Neural Information Processing Systems  
The algorithm is based on branch and bound and integrates dynamic programming for both domain pruning and for obtaining strong bounds for search-space pruning.  ...  Empirically, we show that the approach dominates in terms of running times a recent integer programming approach (and thereby also a recent constraint optimization approach) for the problem.  ...  Theorem 1 states that for any (partial) solution (i.e., an ordered decomposable DAG), there always exists an equivalent solution that does not contain any violations of the preferred vertex order.  ... 
dblp:conf/nips/RantanenHJ17 fatcat:rc43mc26cff73ljd6kmxmy2qfe

Hierarchical Bayesian optimization algorithm: toward a new generation of evolutionary algorithms

2006 ChoiceReviews  
To estimate the distribution, techniques for modeling multivariate data by Bayesian networks are used. The proposed algorithm identi es, reproduces and mixes building blocks up to a speci ed order.  ...  Preliminary experiments show that the BOA outperforms the simple genetic algorithm even on decomposable functions with tight building blocks as a problem size grows.  ...  Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any c o p yright notation thereon.  ... 
doi:10.5860/choice.43-2847 fatcat:w4llps6xwfdmzm7jol7jjxehpq

Page 805 of The Journal of the Operational Research Society Vol. 53, Issue 7 [page]

2002 The Journal of the Operational Research Society  
Consider a set of jobs N = {1 n} in EDD order where each of the jobs decomposes at only one position according to the optimal algorithm.  ...  We then drew comparisons between the MDD rule and the optimal decomposition algorithm and demon- strated that their algorithmic structures have some striking similarities.  ... 

Markov-HTN Planning Approach to Enhance Flexibility of Automatic Web Service Composition

Kun Chen, Jiuyun Xu, Stephan Reiff-Marganiec
2009 2009 IEEE International Conference on Web Services  
Furthermore, an evaluation method to choose the optimal plan and some experimental results illustrate that the proposed approach works effectively.  ...  In the model, HTN planning is enhanced to decompose a task in multiple ways and hence be able to find more than one plan, taking both functional and non-functional properties into account.  ...  Acknowledgment Thanks to the anonymous reviewers for the comments  ... 
doi:10.1109/icws.2009.43 dblp:conf/icws/ChenXR09 fatcat:fxsybxwpcngaja5xjo3hbauv5y

Learning Bayesian networks with ancestral constraints

Eunice Yuh-Jie Chen, Yujia Shen, Arthur Choi, Adnan Darwiche
2016 Neural Information Processing Systems  
Our approach is based on a recently proposed framework for optimal structure learning based on non-decomposable scores, which is general enough to accommodate ancestral constraints.  ...  The proposed framework exploits oracles for learning structures using decomposable scores, which cannot accommodate ancestral constraints since they are non-decomposable.  ...  For example, DAG X → Y → Z expresses the same ancestral relations, after adding edge X → Z.  ... 
dblp:conf/nips/ChenSCD16 fatcat:heded6nw55g6pnrfqn3prvos3u

A Structural SVM Based Approach for Optimizing Partial AUC

Harikrishna Narasimhan, Shivani Agarwal
2013 International Conference on Machine Learning  
optimization problem in the case of partial AUC is harder to decompose.  ...  One of our key technical contributions is an efficient algorithm for solving this combinatorial optimization problem that has the same computational complexity as Joachims' algorithm for optimizing the  ...  Thanks to the anonymous reviewers for helpful comments. HN thanks Microsoft Research India for a partial travel grant to attend the conference.  ... 
dblp:conf/icml/NarasimhanA13 fatcat:ngesxfc7evaqvp236ljsfhiq3e

Learning bayesian networks consistent with the optimal branching

Alexandra M. Carvalho, Arlindo L. Oliveira
2007 Sixth International Conference on Machine Learning and Applications (ICMLA 2007)  
The optimal branching is used as an heuristic for a primary causality order between network variables, which is subsequently refined, according to a certain score, into an optimal CkG Bayesian network.  ...  The proposed algorithm can be applied to scores that decompose over the network structure, such as the well known LL, MDL, AIC, BIC, K2, BD, BDe, BDeu and MIT scores.  ...  This work was partially supported by EU FEDER via FCT project POSC/EIA/ 57398/2004.  ... 
doi:10.1109/icmla.2007.74 dblp:conf/icmla/CarvalhoO07 fatcat:rdp53hwprbf4daj7nezufsezmy

A tutorial on decomposition methods for network utility maximization

D.P. Palomar, Mung Chiang
2006 IEEE Journal on Selected Areas in Communications  
Finally, we present recent examples on: systematic search for alternative decompositions; decoupling techniques for coupled objective functions; and decoupling techniques for coupled constraint sets that  ...  are not readily decomposable.  ...  ACKNOWLEDGMENT The authors would like to thank all the collaborators on the subject of decomposition methods and distributed algorithms for network utility maximization, especially, coauthors M.  ... 
doi:10.1109/jsac.2006.879350 fatcat:yxs5ssvnqngbxpxsolspsnukhq

Fitness-based Linkage Learning in the Real-Valued Gene-pool Optimal Mixing Evolutionary Algorithm

Chantal Olieman, Anton Bouter, Peter A. N. Bosman
2020 IEEE Transactions on Evolutionary Computation  
The recently introduced Real-Valued Gene-pool Optimal Mixing Evolutionary Algorithm (RV-GOMEA) has been shown to be among the state-of-the-art for solving grey-box optimization problems where partial evaluations  ...  a black-box setting where partial evaluations can not be leveraged. 1  ...  Since various real-world optimization problems, as well as all our benchmark problems allow for partial evaluations, we have decided to focus mainly on grey-box optimization in order to obtain a realistic  ... 
doi:10.1109/tevc.2020.3039698 fatcat:brkd3mempracxb47y7cfugsd3y

Linkage Problem, Distribution Estimation, and Bayesian Networks

Martin Pelikan, David E. Goldberg, Erick Cantú-Paz
2000 Evolutionary Computation  
The proposed algorithm is called the Bayesian Optimization Algorithm (BOA).  ...  Except for the maximal order of interactions to be covered, the algorithm does not use any prior knowledge about the problem.  ...  Mengshoel for valuable discussions and useful comments. For help with performing a part of the experiments, the authors would also like to thank Pavel Petrovic.  ... 
doi:10.1162/106365600750078808 pmid:11001554 fatcat:2m23bnimszaolfm6dc62qbp7ei

The Neighbor-Net Algorithm [article]

Dan Levy, Lior Pachter
2008 arXiv   pre-print
The algorithm is optimal for Kalmanson matrices, from which it follows that neighbor-net is consistent and has optimal radius 1/2.  ...  The neighbor-net algorithm is an extension of the neighbor-joining algorithm and is used for constructing split networks.  ...  Equivalently, a partial circular ordering is a partition C of X into ordered sets C = {C 1 , . . . , C m } where each C r ⊆ X and i, j are adjacent elements in C r for some r iff i, j correspond to adjacent  ... 
arXiv:math/0702515v2 fatcat:lqomqg25wrflzfsvknuqqyqike

Decomposition During Search for Propagation-Based Constraint Solvers [article]

Martin Mann and Guido Tack and Sebastian Will
2008 arXiv   pre-print
The presented search algorithm dynamically decomposes sub-problems of a constraint satisfaction problem into independent partial problems, avoiding redundant work.  ...  We have implemented DDS for the Gecode constraint programming library.  ...  Sebastian Will is partially supported by the EU Network of Excellence REWERSE (project number 506779).  ... 
arXiv:0712.2389v2 fatcat:m5m2z7iu4nepfoi5edkqz3qg2e

Hierarchical problem solving with the linkage tree genetic algorithm

Dirk Thierens, Peter A.N. Bosman
2013 Proceeding of the fifteenth annual conference on Genetic and evolutionary computation conference - GECCO '13  
Results show that, although LTGA is a simple algorithm compared to SEAM and hBOA, it nevertheless is a very efficient, reliable, and scalable algorithm for solving the randomly shuffled versions of HIFF  ...  Hierarchical problems represent an important class of nearly decomposable problems. The concept of near decomposability is central to the study of complex systems.  ...  A competent hierarchical optimizer must be capable of representing partial solutions at each level compactly to enable the algorithm to effectively process partial solutions of larger order (this becomes  ... 
doi:10.1145/2463372.2463477 dblp:conf/gecco/ThierensB13 fatcat:cr37g57zqbfnneepjfuzdcxedu

Improved Bagging Algorithm for Pattern Recognition in UHF Signals of Partial Discharges

Tianyan Jiang, Jian Li, Yuanbing Zheng, Caixin Sun
2011 Energies  
This approach establishes the sample information entropy for each sample and the re-sampling process of the traditional Bagging algorithm is optimized.  ...  The optimized third order Peano fractal antenna was applied to capture the PD UHF signals.  ...  The fund of National Basic Research Program of China (973 program, 2009CB724508) and the Natural Science Foundation of Chongqing, China (CSTC 2009BA4048) are also appreciated for supporting this work.  ... 
doi:10.3390/en4071087 fatcat:pnqpkpfokbehzoz5h2wnwfsjwu

Fast Boolean matching based on NPN classification

Zheng Huang, Lingli Wang, Yakov Nasikovskiy, Alan Mishchenko
2013 2013 International Conference on Field-Programmable Technology (FPT)  
This paper proposes a fast algorithm for Boolean matching of completely specified Boolean functions.  ...  The algorithm is conceptually simpler, faster, and more scalable than previous work.  ...  Profiling the algorithm The functions are partitioned into three sets using disjointsupport decomposition [1] [4] : fully decomposable, partially decomposable, and non-decomposable.  ... 
doi:10.1109/fpt.2013.6718374 dblp:conf/fpt/HuangWNM13 fatcat:wyzhft3zijd2vozkux3lp6qz24
« Previous Showing results 1 — 15 out of 143,176 results