• Add four surrogate-assisted multi-objective evolutionary algorithms AB-SAEA, EDN-ARMOEA, HeE-MOEA, KTA2, and a constrained multi-objective evolutionary algorithm c-DPEA. There are currently 158 algorithms in the platform.

• Add two multi-objective optimization algorithms CCGDE3 and NSGA-II+ARSBX and one single-objective optimization algorithm OFA. There are currently 153 algorithms in the platform.

• Fix some minor bugs in algorithms and the GUI.

• 20+ algorithms and 100+ problems for single-objective optimization. There are currently 150 algorithms and 339 problems in the platform, including single-objective optimization, multi-objective optimization, many-objective optimization, combinatorial optimization, large-scale optimization, constrained optimization, multimodal optimization, expensive optimization, sparse optimization, and preference optimization.

• A totally new GUI with more powerful functions, which contains a test module, an application module, and an experiment module.

• A novel filter system based on hybrid labels, which facilitates the selection of suitable algorithms for solving different types of problems.

• More convenient interfaces for solving user-defined problems, where no file needs to be written by users.

• A better visualization of populations, where the true Pareto fronts and feasible regions can be shown in the plots.

• Add one algorithm for constrained optimization (i.e., CMOEA-MS), one algorithm for large-scale optimization (i.e., DGEA), one algorithm for expensive optimization (i.e., MESMO), and one algorithm for feature selection (i.e., DAEA). There are currently 122 algorithms in the platform.

• Add three algorithms for constrained optimization (i.e., CCMO, MOEA/D-DAE, and TiGE-2) and an algorithm for many-objective optimization (i.e., PREA). There are currently 118 algorithms in the platform.

• Fix some minor bugs in the Pareto front sampling methods in LIR-CMOP and MW problems.

• Add the time-varying ratio error estimation (TREE) test suite, which contains six constrained large-scale problems from real-world applications.
• Fix some minor bugs in algorithms and problems.

Release Highlights of PlatEMO 2.3

• Add four algorithms: C-TAEA, ToP, MOEA/D-URAW, and MultiObjectiveEGO. There are currently 108 algorithms on the platform.
• Add the constrained benchmark problems DOC1-9 and MW1-14. There are currently 201 problems on the platform.
• Update the Pareto front sampling methods of DAS-CMOP1-9 and LIR-CMOP1-14: Dynamically sample points on Pareto fronts instead of loading points from files.
• Update the table in the experiment module: Ignore NaN values when calculating the mean and standard deviation in each cell of the table.

Copyright

The Copyright of the PlatEMO belongs to the BIMK group. You are free to use the PlatEMO for research purposes. All publications which use this platform or any code in the platform should acknowledge the use of "PlatEMO" and reference "Ye Tian, Ran Cheng, Xingyi Zhang, and Yaochu Jin, PlatEMO: A MATLAB Platform for Evolutionary Multi-Objective Optimization [Educational Forum], IEEE Computational Intelligence Magazine, 2017, 12(4): 73-87".

@article{PlatEMO,
  title={{PlatEMO}: A {MATLAB} platform for evolutionary multi-objective optimization},
  author={Tian, Ye and Cheng, Ran and Zhang, Xingyi and Jin, Yaochu},
  journal={IEEE Computational Intelligence Magazine},
  volume={12},
  number={4},
  pages={73--87},
  year={2017},
}

fix a high-level bug.

Release Highlights of PlatEMO 2.2

• Add two algorithms AGE-MOEA and PPS.
• Add the constrained benchmark problems DAS-CMOP1-9 and LIR-CMOP1-14.

Note

Add the sparse multi-objective evolutionary algorithm SparseEA. Add the sparse multi-objective test suite SMOP1-SMOP8. Add four sparse multi-objective optimization problems, i.e., feature selection, pattern mining, critical node detection, and neural network training. Add the diversity metric CPF (i.e., coverage over Pareto front). Add the irregular multi-objective test suite IMOP1-IMOP8.

Copyright

The Copyright of the PlatEMO belongs to the BIMK group. You are free to use the PlatEMO for research purposes. All publications which use this platform or any code in the platform should acknowledge the use of "PlatEMO" and reference "Ye Tian, Ran Cheng, Xingyi Zhang, and Yaochu Jin, PlatEMO: A MATLAB Platform for Evolutionary Multi-Objective Optimization [Educational Forum], IEEE Computational Intelligence Magazine, 2017, 12(4): 73-87".

@article{PlatEMO,
  title={{PlatEMO}: A {MATLAB} platform for evolutionary multi-objective optimization},
  author={Tian, Ye and Cheng, Ran and Zhang, Xingyi and Jin, Yaochu},
  journal={IEEE Computational Intelligence Magazine},
  volume={12},
  number={4},
  pages={73--87},
  year={2017},
}