A Tutorial on the Performance Assessment of Stochastic Multiobjective Optimizers

By Joshua D. Knowles, Lothar Thiele, Eckart Zitzler. 2006

876 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics

View Paper: https://www.semanticscholar.org/paper/877d158ac1e5d77623359e10e62b244d574af8a6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

Citations in the corpus

Empirical evaluation methods for multiobjective reinforcement learning algorithms (2011, background): “More importantly these metrics can prove misleading, as they may indicate a front to be superior to another front when the latter actually weakly dominates the former (Knowles et al. 2006).”
Multiobjective Immune Algorithm with Nondominated Neighbor-Based Selection (2008, background): “Zitzler et al. (2003) and Knowles et al. (2006) have suggested that the power of unary quality indicators was restricted.”
Automatically improving the anytime behaviour of optimisation algorithms (2014, methodology): “…automatic configuration , the available Kolmogorov-Smirnov test based on the attainment function (Knowles, Thiele, & Zitzler, 2006) is not enough, and graphical exploration methods (Knowles et al., 2006; Knowles, 2005; López-Ibáñez, Paquete, & Stützle, 2010) require substantial human interaction.”
Approximating the Set of Pareto-Optimal Solutions in Both the Decision and Objective Spaces by an Estimation of Distribution Algorithm (2009, methodology): “Performance Metrics The inverted generational distance (IGD) metric [14], [31] and hypervolume difference (I − H ) metric [32] are used to assess the algorithm performances in our experimental studies.”
Fuzzy-Based Pareto Optimality for Many-Objective Evolutionary Algorithms (2014, background): “can be rejected in favor of an alternative test H1 [28].”
The Hypervolume Indicator Revisited: On the Design of Pareto-compliant Indicators Via Weighted Integration (2007, background): “One important feature an indicator should have is Pareto compliance [9], i.”
Constraint Handling in Multiobjective Evolutionary Optimization (2009, methodology): “The quantitative comparisons are illustrated by statistical box plots, and a Mann–Whitney rank-sum test is implemented to evaluate whether the difference in performance between two independent samples is significant [40].”
Quality Assessment of Pareto Set Approximations (2008, methodology): “Details for this statistical testing procedure are given in (Knowles et al., 2006).”
Modified Distance Calculation in Generational Distance and Inverted Generational Distance (2015, background): “Various performance indicators have been proposed to evaluate the quality of a non-dominated solution set [8], [14], [15], [26].”
The jMetal framework for multi-objective optimization: Design and architecture (2010, background): “A number of quality indicators have been proposed in the literature: Generational Distance (GD) [39], Inverse Gene rational Distance (IGD), Hypervolume (HV) [40], Epsilon [41] , Spread or∆ [2], Generalized Spread indicators, and many others.”
jMetal: A Java framework for multi-objective optimization (2011, methodology): “A number of quality indicators for measuring these two criteria have been proposed in the literature: Generational Distance (GD) [39], Inverse Generational Distance (IGD), Hypervolume (HV) [38], Epsilon [40], Spread or D [4], Generalized Spread indicators, and others.”
SMPSO: A new PSO-based metaheuristic for multi-objective optimization (2009, methodology): “For assessing the performance of the algorithms, we have considered three quality indicators: additive unary epsilon indicator ( I 1 (cid:2) + ) [9], spread ( Δ ) [3], and hypervolume (HV) [18].”
Covariance Matrix Adaptation for Multi-objective Optimization (2007, methodology): “We concisely define the performance measures used, for a detailed description of the methods we refer to the literature (Knowles and Corne, 2002; Zitzler et al., 2003; Knowles et al., 2005).”