Stochastic ranking for constrained evolutionary optimization

By Thomas Philip Runarsson, Xin Yao. IEEE Transactions on Evolutionary Computation, 2000

1,807 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

View Paper: https://www.semanticscholar.org/paper/827e3d01e52438e29aac1f6c4c8a2ed972a87c75

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Theory and Mathematics

Citations in the corpus

An improved epsilon constraint-handling method in MOEA/D for CMOPs with large infeasible regions (2017, methodology): “MOEA/D-SR (Jan and Khanum (2013)) embeds the stochastic ranking method (SR) (Runarsson and Yao (2000)) in MOEA/D to deal with constraints.”
An improved particle swarm optimizer for mechanical design optimization problems (2004, methodology): “Best solution found Design variables PSO Runarsson [27] GRG [25] Gen [26]”
A generic framework for constrained optimization using genetic algorithms (2005, methodology): “In [34], the authors characterize the problem of choosing the appropriate penalty coefficient for each constraint and describe how it affects the domination between the constraint violation and the objective function in deciding the rank of each individual.”
Hybridizing particle swarm optimization with differential evolution for constrained numerical and engineering optimization (2010, methodology): “Runarsson and Yao [7] introduced a stochastic ranking method to balance the objective and penalty functions.”
Self-adaptive fitness formulation for constrained optimization (2003, methodology): “A comparison of the algorithm performance is also made to the results obtained by Runarsson and Yao [23] and Ben Hamida and Schoenauer [1] for the same test problems although it is understood that both methods used an evolution strategy algorithm and performed different experiments to those given in”
DEPSO: hybrid particle swarm with differential evolution operator (2003, background): “For G3, G5, G11, which have almost 0% feasible space (SF) due to the equality constraints, are needed to replace the constraint ( ) 0 g x = r by an inequality constraint | ( ) | g x ε < r for some small ε >0 [13, 14, 17].”
A Survey on Evolutionary Constrained Multiobjective Optimization (2023, methodology): “These methods compare objectives and constraints separately, mainly including the CDP [45], ε constrained method [50], [51], and stochastic ranking (SR) [52].”
Bayesian Optimization for Adaptive Experimental Design: A Review (2020, methodology): “For more information, see http://creativecommons.org/licenses/by/4.0/ 13937 multi-start derivative free local optimiser e.g. COBYLA [36], or evolutionary algorithms e.g. ISRES [37], or Lipschitzian methods such as DIRECT [34].”
Well placement optimization subject to realistic field development constraints (2016, methodology): “Multi-objective optimization technique In this technique, the constrained optimization problem is considered as a multi-objective optimization problem in which both the objective function and the constraint violations are minimized [35, 36].”
A ranking selection-based particle swarm optimizer for engineering design optimization problems (2008, methodology): “... bounds, not approximately like other approaches).The statistical results of these accurately feasible solutions are listed in the row labeled “RSPSO (1)” of Table 11. However, to compare RSPSO with other approaches, we also conducted an experiment where the equality constraints were converted in”
Push and Pull Search for Solving Constrained Multi-objective Optimization Problems (2017, methodology): “SR [12] is different from CDP and EC in terms of the order of comparison.”
Constraint-handling in genetic algorithms through the use of dominance-based tournament selection (2002, methodology): “The best solution of our approach is compared in Table 4. Note that the approach proposed by Koziel & Michalewicz [22] required 1,400,000 evaluations of the fitness function to produce the result shown in Table 4. The approach of Runarsson & Yao [ 31 ] required 350,000 evaluations of the fitness fun”