Adaptive threshold parameter estimation with recursive differential grouping for problem decomposition

Sun, Yuan; Omidvar, Mohammad Nabi; Kirley, Michael; Li, Xiaodong

doi:10.1145/3205455.3205483

Cited by 52 publications

(24 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, one global threshold value may not be sufficient for determining interactions of variables in subcomponents having a dissimilar contribution to the fitness value. To solve these problems, a decomposition method inspired by DG2 and RDG was proposed, called RDG2 [49]. In RDG2, the threshold value is adaptively estimated based on the computational round-off errors of RDG.…”

Section: Review On Problem Decomposition Approaches Using CCmentioning

confidence: 99%

“…By exploring the interdependency information in the problem, a number of detection-based static decomposition techniques proposed to make the near-optimal decomposition. Examples of such static decomposition methods include CCVIL [31], DG [44], XDG [45], GDG [46], RDG [10], D-GDG [48], RDG2 [49], and RDG3 [50]. These methods are also called automatic decomposition strategies as they undertake the variable interactions to group the variables [10].…”

Section: Review On Problem Decomposition Approaches Using CCmentioning

confidence: 99%

See 1 more Smart Citation

Cooperative co-evolution for feature selection in Big Data with random feature grouping

et al. 2020

View full text Add to dashboard Cite

A massive amount of data is generated with the evolution of modern technologies. This high-throughput data generation results in Big Data, which consist of many features (attributes). However, irrelevant features may degrade the classification performance of machine learning (ML) algorithms. Feature selection (FS) is a technique used to select a subset of relevant features that represent the dataset. Evolutionary algorithms (EAs) are widely used search strategies in this domain. A variant of EAs, called cooperative co-evolution (CC), which uses a divide-and-conquer approach, is a good choice for optimization problems. The existing solutions have poor performance because of some limitations, such as not considering feature interactions, dealing with only an even number of features, and decomposing the dataset statically. In this paper, a novel random feature grouping (RFG) has been introduced with its three variants to dynamically decompose Big Data datasets and to ensure the probability of grouping interacting features into the same subcomponent. RFG can be used in CC-based FS processes, hence called Cooperative Co-Evolutionary-Based Feature Selection with Random Feature Grouping (CCFSRFG). Experiment analysis was performed using six widely used ML classifiers on seven different datasets from the UCI ML repository and Princeton University Genomics repository with and without FS. The experimental results indicate that in most cases [i.e., with naïve Bayes (NB), support vector machine (SVM), k-Nearest Neighbor (k-NN), J48, and random forest (RF)] the proposed CCFSRFG-1 outperforms an existing solution (a CC-based FS, called CCEAFS) and CCFSRFG-2, and also when using all features in terms of accuracy, sensitivity, and specificity.

show abstract

Section: Review On Problem Decomposition Approaches Using CCmentioning

confidence: 99%

Section: Review On Problem Decomposition Approaches Using CCmentioning

confidence: 99%

Cooperative co-evolution for feature selection in Big Data with random feature grouping

et al. 2020

View full text Add to dashboard Cite

show abstract

“…However in practice, the value of λ for separable decision variables may be non-zero, due to the computational round-off errors incurred by the floating-point operations [15]. In [39], we applied the technique suggested by DG2 [15] to estimate an upper bound on the round-off errors associated with the calculation of the non-linearity term λ: Here…”

Section: Recursive Differential Groupingmentioning

confidence: 99%

“…where µ M is a machine dependent constant. The upper bound is then used as the threshold value in RDG2 [39] to distinguish between separable and non-separable variables: With Theorem 1, the interaction between two subsets of decision variables (X 1 and X 2 ) can be identified by the following procedure: 1) Set all the decision variables to the lower bounds (lb) of the search space (x l,l ); 2) Perturb the decision variables X 1 of x l,l from the lower bounds to the upper bounds (ub), denoted as x u,l ; 3) Calculate the fitness change ∆ 1 between x l,l and x u,l ; 4) Perturb decision variables X 2 of x l,l (x u,l ) from lb to the middle of the search space, denoted as x l,m (x u,m ); 5) Calculate the fitness change ∆ 2 between x l,m and x u,m ; 6) If the difference (λ) between ∆ 1 and ∆ 2 is greater than the threshold ǫ, X 1 and X 2 interact.…”

Section: Recursive Differential Groupingmentioning

confidence: 99%