Effect of size and order of variables in rules for multi-objective repair-based innovization procedure

Innovization is a task of learning common principles that exist among some or all of the Pareto-optimal solutions of a multi-objective optimization problem. Except a few earlier studies, most innovization related studies were performed on the final non-dominated solutions found by an EMO algorithm. Recently, authors showed that these principles can be learned during an optimization run and simultaneously utilized in the same optimization run to repair variables to achieve a faster convergence to the Pareto-optimal set. Different principles learned during an optimization run can not only have different number of variables but, may also have variables that are common among a number of principles. Moreover, a preference order for repairing variables may play an important role for proper convergence. Thus, when multiple principles exist, it is important to use a strategy that is most beneficial for repairing evolving population of solutions. This paper makes a first attempt to assess and understand the effect of different strategies to make innovization-based repair of variables most useful. Based on results on test problems, the paper also makes useful suggestions, which require immediate further experimentation on more complex and real-world problems.

[1]  Kalyanmoy Deb,et al.  Interleaving Innovization with Evolutionary Multi-Objective Optimization in Production System Simulation for Faster Convergence , 2013, LION.

[2]  M. Newman Power laws, Pareto distributions and Zipf's law , 2004, cond-mat/0412004.

[3]  Lothar Thiele,et al.  Comparison of Multiobjective Evolutionary Algorithms: Empirical Results , 2000, Evolutionary Computation.

[4]  Mohamed Wiem Mkaouer,et al.  Recommendation system for software refactoring using innovization and interactive dynamic optimization , 2014, ASE.

[5]  Kalyanmoy Deb,et al.  Data mining methods for knowledge discovery in multi-objective optimization: Part A - Survey , 2017, Expert Syst. Appl..

[6]  DebK.,et al.  A fast and elitist multiobjective genetic algorithm , 2002 .

[7]  Kalyanmoy Deb,et al.  Hybrid evolutionary multi-objective optimization and analysis of machining operations , 2012 .

[8]  Kalyanmoy Deb,et al.  Automated discovery of vital knowledge from Pareto-optimal solutions: First results from engineering design , 2010, IEEE Congress on Evolutionary Computation.

[9]  A. C. Rencher Methods of multivariate analysis , 1995 .

[10]  Kalyanmoy Deb,et al.  Simulation-based Innovization for production systems improvement : An industrial case study , 2009 .

[11]  Kalyanmoy Deb,et al.  An integrated approach to automated innovization for discovering useful design principles: Case studies from engineering , 2014, Appl. Soft Comput..

[12]  Kalyanmoy Deb,et al.  Adaptive Use of Innovization Principles for a Faster Convergence of Evolutionary Multi-Objective Optimization Algorithms , 2016, GECCO.

[13]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[14]  Hyunseung Bang,et al.  Extracting and Applying Knowledge with Adaptive Knowledge-Driven Optimization to Architect an Earth Observing Satellite System , 2017 .

[15]  Aravind Srinivasan,et al.  Innovization: innovating design principles through optimization , 2006, GECCO.