What question did this study set out to answer?

To develop an evolutionary algorithm tailored for large-scale multiobjective optimization problems where the majority of decision variables in optimal solutions are zero.

May 22, 2019IEEE Transactions on Evolutionary Computation

An Evolutionary Algorithm for Large-Scale Sparse Multiobjective Optimization Problems

View Full Paper

Ask AI

Bookmark

Authors

YTYe TianUniversity of Southern California XZXingyi ZhangUniversity of Bern CWChao WangShanghai University

Discussion

Member takes

Implication

Algorithm development study demonstrates superior search performance across large-scale sparse multiobjective problems, highlighting improved efficiency for high-dimensional applications.

Key Points

To develop an evolutionary algorithm tailored for large-scale multiobjective optimization problems where the majority of decision variables in optimal solutions are zero.
Designed an evolutionary algorithm incorporating a novel population initialization strategy and genetic operators customized to preserve solution sparsity.
Constructed a dedicated benchmark test suite to evaluate algorithm behavior on large-scale sparse optimization scenarios.
Benchmarked the proposed algorithm against seven existing evolutionary algorithms across the test suite and four real-world engineering applications, including feature selection and neural network structure optimization.
Demonstrated superior optimization performance over all seven comparative evolutionary algorithms on the synthetic test suite.
Outperformed benchmark algorithms across all four real-world application test cases while effectively maintaining decision-variable sparsity.

Cite This Study

Tian et al. (2019) studied this question.

synapsesocial.com/papers/6a0ec8f5a14f152feaf9d6bc https://doi.org/10.1109/tevc.2019.2918140

View Full Paper

Ask AI

Bookmark

Implication

Algorithm development study demonstrates superior search performance across large-scale sparse multiobjective problems, highlighting improved efficiency for high-dimensional applications.

An Evolutionary Algorithm for Large-Scale Sparse Multiobjective Optimization Problems

Discussion

Member takes

Implication

Key Points

Cite This Study

An Evolutionary Algorithm for Large-Scale Sparse Multiobjective Optimization Problems

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider