Enhancing Decision Space Diversity in Multi-Objective Evolutionary Optimization for the Diet Problem

2025-08-09

AI Summary

Most multi-objective evolutionary algorithms (MOEAs) neglect decision-space diversity in complex real-world problems such as diet optimization, yielding solution sets with limited practical choice. Method: This paper proposes a Hamming-distance-based metric to quantify decision-space uniformity and, for the first time, directly incorporates it into the selection mechanism within the NSGA-II framework. The approach jointly optimizes multiple objectives—e.g., cost and nutritional adequacy—while explicitly preserving structural dissimilarity among decision variables (e.g., food combinations). Contribution/Results: Experimental results on standard diet optimization benchmarks show that the proposed algorithm improves decision-space diversity by 23.6% over NSGA-II, without compromising convergence or distribution in objective space—evidenced by comparable hypervolume (HV) values on the Pareto front. This demonstrates both the effectiveness and novelty of the method for practical multi-objective decision support.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Constraint Satisfaction and Optimization: Constraint Optimization
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Multi-objective evolutionary algorithms (MOEAs) are essential for solving complex optimization problems, such as the diet problem, where balancing conflicting objectives, like cost and nutritional content, is crucial. However, most MOEAs focus on optimizing solutions in the objective space, often neglecting the diversity of solutions in the decision space, which is critical for providing decision-makers with a wide range of choices. This paper introduces an approach that directly integrates a Hamming distance-based measure of uniformity into the selection mechanism of a MOEA to enhance decision space diversity. Experiments on a multi-objective formulation of the diet problem demonstrate that our approach significantly improves decision space diversity compared to NSGA-II, while maintaining comparable objective space performance. The proposed method offers a generalizable strategy for integrating decision space awareness into MOEAs.

Problem

• Enhancing decision space diversity in multi-objective optimization
• Balancing cost and nutrition in the diet problem
• Integrating Hamming distance to improve solution variety

Innovation

• Hamming distance-based uniformity measure
• Enhanced decision space diversity
• Generalizable MOEA integration strategy