Merge branch 'dev'

Author: geoffreyp

moead_framework/test/test_genetic_mating.py

@@ -0,0 +1,121 @@
+
+@article{moead_de,
+  title = {{{MOEA}}/{{D}}-{{DE}} : {{Multiobjective Optimization Problems With Complicated Pareto Sets}}, {{MOEA}}/{{D}} and {{NSGA}}-{{II}}},
+  author = {Li, H. and Zhang, Q.},
+  year = {2009},
+  month = apr,
+  volume = {13},
+  pages = {284--302},
+  journal = {IEEE Transactions on Evolutionary Computation},
+  keywords = {Aggregation,Computer science,decomposition,differential evolution,evolutionary algorithms,evolutionary computation,Evolutionary computation,MOEA/D,multiobjective optimization,multiobjective optimization problems,NSGA-II,Pareto optimality,Pareto optimisation,Pareto optimization,Pareto sets,Performance evaluation,set theory,Shape control,test problems,Testing},
+  number = {2}
+}
+
+@article{moead,
+  title = {{{MOEA}}/{{D}}: {{A Multiobjective Evolutionary Algorithm Based}} on {{Decomposition}}},
+  shorttitle = {{{MOEA}}/{{D}}},
+  author = {{Qingfu Zhang} and {Hui Li}},
+  year = {2007},
+  volume = {11},
+  pages = {712--731},
+  journal = {IEEE Transactions on Evolutionary Computation},
+  language = {en},
+  number = {6}
+}
+
+@inproceedings{moead_dra,
+  title = {{{MOEA}}/{{D}}-{{DRA}}: {{The}} Performance of a New Version of {{MOEA}}/{{D}} on {{CEC09}} Unconstrained {{MOP}} Test Instances},
+  booktitle = {2009 {{IEEE Congress}} on {{Evolutionary Computation}}},
+  author = {Zhang, Qingfu and Liu, Wudong and Li, Hui},
+  year = {2009},
+  month = may,
+  pages = {203--208},
+  publisher = {{IEEE}},
+  address = {{Trondheim, Norway}},
+  language = {en}
+}
+
+@inproceedings{gpruvost_gecco2020,
+author = {Pruvost, Geoffrey and Derbel, Bilel and Liefooghe, Arnaud and Verel, S\'{e}bastien and Zhang, Qingfu},
+title = {Surrogate-Assisted Multi-Objective Combinatorial Optimization Based on Decomposition and Walsh Basis},
+year = {2020},
+isbn = {9781450371285},
+publisher = {Association for Computing Machinery},
+address = {New York, NY, USA},
+url = {https://doi.org/10.1145/3377930.3390149},
+doi = {10.1145/3377930.3390149},
+booktitle = {Proceedings of the 2020 Genetic and Evolutionary Computation Conference},
+pages = {542–550},
+numpages = {9},
+keywords = {discrete surrogates, decomposition, multi-objective optimization},
+location = {Canc\'{u}n, Mexico},
+series = {GECCO ’20}
+}
+
+@InProceedings{gpruvost_evocop2020,
+author="Pruvost, Geoffrey and Derbel, Bilel and Liefooghe, Arnaud and Li, Ke and Zhang, Qingfu",
+editor="Paquete, Lu{\'i}s
+and Zarges, Christine",
+title="On the Combined Impact of Population Size and Sub-problem Selection in MOEA/D",
+booktitle="Evolutionary Computation in Combinatorial Optimization",
+year="2020",
+publisher="Springer International Publishing",
+address="Cham",
+pages="131--147",
+abstract="This paper intends to understand and to improve the working principle of decomposition-based multi-objective evolutionary algorithms. We review the design of the well-established Moea/d framework to support the smooth integration of different strategies for sub-problem selection, while emphasizing the role of the population size and of the number of offspring created at each generation. By conducting a comprehensive empirical analysis on a wide range of multi- and many-objective combinatorial NK landscapes, we provide new insights into the combined effect of those parameters on the anytime performance of the underlying search process. In particular, we show that even a simple random strategy selecting sub-problems at random outperforms existing sophisticated strategies. We also study the sensitivity of such strategies with respect to the ruggedness and the objective space dimension of the target problem.",
+isbn="978-3-030-43680-3"
+}
+
+@article{Campelo_2020,
+   title={The MOEADr Package: A Component-Based Framework for Multiobjective Evolutionary Algorithms Based on Decomposition},
+   volume={92},
+   ISSN={1548-7660},
+   url={http://dx.doi.org/10.18637/jss.v092.i06},
+   DOI={10.18637/jss.v092.i06},
+   number={6},
+   journal={Journal of Statistical Software},
+   publisher={Foundation for Open Access Statistic},
+   author={Campelo, Felipe and Batista, Lucas S. and Aranha, Claus},
+   year={2020}
+}
+
+@ARTICLE{pymoo,
+    author={J. {Blank} and K. {Deb}},
+    journal={IEEE Access},
+    title={Pymoo: Multi-Objective Optimization in Python},
+    year={2020},
+    volume={8},
+    number={},
+    pages={89497-89509},
+}
+
+@article{pygmo,
+  doi = {10.21105/joss.02338},
+  url = {https://doi.org/10.21105/joss.02338},
+  year = {2020},
+  publisher = {The Open Journal},
+  volume = {5},
+  number = {53},
+  pages = {2338},
+  author = {Francesco Biscani and Dario Izzo},
+  title = {A parallel global multiobjective framework for optimization: pagmo},
+  journal = {Journal of Open Source Software}
+}
+
+@inproceedings{jmetal,
+author = {Nebro, Antonio J. and Durillo, Juan J. and Vergne, Matthieu},
+title = {Redesigning the JMetal Multi-Objective Optimization Framework},
+year = {2015},
+isbn = {9781450334884},
+publisher = {Association for Computing Machinery},
+address = {New York, NY, USA},
+url = {https://doi.org/10.1145/2739482.2768462},
+doi = {10.1145/2739482.2768462},
+abstract = {jMetal, an open source, Java-based framework for multi-objective optimization with metaheuristics, has become a valuable tool for many researches in the area as well as for some industrial partners in the last ten years. Our experience using and maintaining it during that time, as well as the received comments and suggestions, have helped us improve the jMetal design and identify significant features to incorporate. This paper revisits the jMetal architecture, describing its refined new design, which relies on design patterns, principles from object-oriented design, and a better use of the Java language features to improve the quality of the code, without disregarding jMetal ever goals of simplicity, facility of use, flexibility, extensibility and portability. Among the newly incorporated features, jMetal supports live interaction with running algorithms and parallel execution of algorithms.},
+booktitle = {Proceedings of the Companion Publication of the 2015 Annual Conference on Genetic and Evolutionary Computation},
+pages = {1093–1100},
+numpages = {8},
+keywords = {open source, multi-objective metaheu- ristics, optimization framework, jmetal},
+location = {Madrid, Spain},
+series = {GECCO Companion '15}
+}

moead_framework/test/test_genetic_selector.py

@@ -0,0 +1,66 @@
+---
+title: 'A modular MOEA/D python framework'
+tags:
+  - Python
+  - optimization
+  - Multi-objective
+  - moead
+  - framework
+authors:
+  - name: Geoffrey Pruvost
+    affiliation: "1, 2" 
+  - name: Bilel Derbel
+    affiliation: "1, 2" 
+  - name: Arnaud Liefooghe
+    affiliation: "1, 2" 
+affiliations:
+ - name: Univ. Lille, CNRS, Centrale Lille, UMR 9189 CRIStAL, F-59000 Lille, France
+   index: 1
+ - name: Inria Lille - Nord Europe
+   index: 2
+
+date: 1 December 2020
+bibliography: paper.bib
+
+# Optional fields if submitting to a AAS journal too, see this blog post:
+# https://blog.joss.theoj.org/2018/12/a-new-collaboration-with-aas-publishing
+# aas-doi: 10.3847/xxxxx <- update this with the DOI from AAS once you know it.
+# aas-journal: Astrophysical Journal <- The name of the AAS journal.
+---
+
+# Summary
+
+MOEA/D (Multi-objective evolutionary algorithm based on decomposition) is a general-purpose 
+algorithm framework. It decomposes a multi-objective optimization problem into a number of 
+single-objective optimization sub-problems and then uses a search heuristic to optimize these 
+sub-problems simultaneously and cooperatively. MOEA/D is a state-of-art framework in aggregation-based 
+approaches for multi-objective optimization.
+
+The goal of this python package is to provide a modular framework for researchers who 
+are working on variants of MOEA/D.
+
+
+# Motivation
+
+The first version of MOEA/D [@moead] and its most famous variants [@moead_de; @moead_dra] are implemented in recent multi-objective 
+optimization software such as pymoo [@pymoo], pygmo [@pygmo] and jMetal [@jmetal]. These software bring a lot of algorithms and problems but 
+don't allow us to test and analyse in detail each algorithms. 
+The modular R package MOEADr [@Campelo_2020] allows to use each components of MOEA/D with some variants but is not enough flexible 
+to add, remove or update the behavior of these components. 
+
+With this python package, we want to bring the modularity of the MOEADr package by using the flexibility of Python to 
+allow users to update the behavior of MOEA/D components in their research works to propose new variants without 
+be limited by the software. For example, the paper [@gpruvost_evocop2020] uses this package to add a new component 
+in MOEA/D (the sub-problem selection strategy) and this paper [@gpruvost_gecco2020] uses 
+the package to rewrite the offspring_generator component to implement a new framework to use surrogate models 
+in combinatorial optimization with multi-objective optimization and decomposition technics.
+
+
+
+# Acknowledgements
+
+This work was supported by the French national research agency(ANR-16-CE23-0013-01) 
+and the Research Grants Council of HongKong (RGC Project No. A-CityU101/16).
+
+
+# References

paper.bib

@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="moead-framework",
-    version="0.5.5",
+    version="0.5.6",
     author="Geoffrey Pruvost",
     author_email="geoffrey@pruvost.xyz",
     description="MOEA/D Framework in Python 3",