Analyzing the Landscape of the Indicator-based Subset Selection Problem

2025-04-11

AI Summary

This work investigates the fitness landscape characteristics of the Indicator-based Subset Selection Problem (ISSP), aiming to uncover how quality indicator types and Pareto front geometry jointly shape landscape structure. Methodologically, it introduces the first systematic integration of classical landscape analysis—such as fitness–distance correlation and ruggedness—with precise Local Optima Networks (LONs) to construct an interpretable ISSP landscape characterization framework. Key findings reveal that the ε-indicator induces pronounced neutrality plateaus alongside high-density multimodal local optima, highlighting the landscape’s extreme sensitivity to both indicator choice and front geometry. These results provide theoretical foundations for algorithm design in subset selection and advance the interpretability of ISSP performance analysis.

Technology Category

• Search and Optimization: Local Search
• Constraint Satisfaction and Optimization: Satisfiability
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

The indicator-based subset selection problem (ISSP) involves finding a point subset that minimizes or maximizes a quality indicator. The ISSP is frequently found in evolutionary multi-objective optimization (EMO). An in-depth understanding of the landscape of the ISSP could be helpful in developing efficient subset selection methods and explaining their performance. However, the landscape of the ISSP is poorly understood. To address this issue, this paper analyzes the landscape of the ISSP by using various traditional landscape analysis measures and exact local optima networks (LONs). This paper mainly investigates how the landscape of the ISSP is influenced by the choice of a quality indicator and the shape of the Pareto front. Our findings provide insightful information about the ISSP. For example, high neutrality and many local optima are observed in the results for ISSP instances with the additive $epsilon$-indicator.

Problem

• Understanding ISSP landscape for better subset selection methods
• Analyzing ISSP impact from quality indicators and Pareto front
• Exploring high neutrality and local optima in ISSP instances

Innovation

• Uses traditional landscape analysis measures
• Employs exact local optima networks (LONs)
• Investigates quality indicator and Pareto front