Multi-objective Optimization with Unbounded Solution Sets
نویسندگان
چکیده
Machine learning often requires the optimization of multiple, partially conflicting objectives. True multi-objective optimization (MOO) methods avoid the need to choose a weighting of the objectives a priori and provide insights about the tradeoffs between the objectives. We extend a state-of-the-art derivative-free Monte Carlo method for MOO, the MO-CMA-ES, to operate on an unbounded set of (non-dominated) candidate solutions. The resulting algorithm, UP-MO-CMA-ES, performed well in two recent performance comparisons of MOO methods.
منابع مشابه
UNBOUNDEDNESS IN MOILP AND ITS EFFICIENT SOLUTIONS
In this paper we investigate Multi-Objective Integer Linear Programming (MOILP) problems with unbounded feasible region and introduce recession direction for MOILP problems. Then we present necessary and sufficient conditions to have unbounded feasible region and infinite optimal values for objective functions of MOILP problems. Finally we present some examples with unbounded feasible region and fi...
متن کاملSolving Critical Path Problem in Project Network by a New Enhanced Multi-objective Optimization of Simple Ratio Analysis Approach with Interval Type-2 Fuzzy Sets
Decision making is an important issue in business and project management that assists finding the optimal alternative from a number of feasible alternatives. Decision making requires adequate consideration of uncertainty in projects. In this paper, in order to address uncertainty of project environments, interval type-2 fuzzy sets (IT2FSs) are used. In other words, the rating of each alternativ...
متن کاملMMDT: Multi-Objective Memetic Rule Learning from Decision Tree
In this article, a Multi-Objective Memetic Algorithm (MA) for rule learning is proposed. Prediction accuracy and interpretation are two measures that conflict with each other. In this approach, we consider accuracy and interpretation of rules sets. Additionally, individual classifiers face other problems such as huge sizes, high dimensionality and imbalance classes’ distribution data sets. This...
متن کاملMULTI-OBJECTIVE OPTIMIZATION WITH PREEMPTIVE PRIORITY SUBJECT TO FUZZY RELATION EQUATION CONSTRAINTS
This paper studies a new multi-objective fuzzy optimization prob- lem. The objective function of this study has dierent levels. Therefore, a suitable optimized solution for this problem would be an optimized solution with preemptive priority. Since, the feasible domain is non-convex; the tra- ditional methods cannot be applied. We study this problem and determine some special structures related...
متن کاملA FAST FUZZY-TUNED MULTI-OBJECTIVE OPTIMIZATION FOR SIZING PROBLEMS
The most recent approaches of multi-objective optimization constitute application of meta-heuristic algorithms for which, parameter tuning is still a challenge. The present work hybridizes swarm intelligence with fuzzy operators to extend crisp values of the main control parameters into especial fuzzy sets that are constructed based on a number of prescribed facts. Such parameter-less particle ...
متن کامل