研究目的
To present a comprehensive review of the existing visualization methods used in evolutionary multiobjective optimization and propose a new visualization method that uses prosection to visualize 4-D approximation sets.
研究成果
The prosection method provides an intuitive visualization of high-dimensional approximation sets, preserving the shape, range, and distribution of vectors, as well as the Pareto dominance relation and relative closeness to reference points for some vectors. It is robust and computationally inexpensive but limited to 4-D sets.
研究不足
The prosection method is limited to visualizing 4-D approximation sets and does not easily scale to higher dimensions.
1：Experimental Design and Method Selection:
The paper reviews existing visualization methods and proposes a new method using prosection for visualizing 4-D approximation sets.
2：Sample Selection and Data Sources:
Two novel 4-D benchmark approximation sets are used to demonstrate the outcomes of the visualization methods.
3：List of Experimental Equipment and Materials:
Not explicitly mentioned.
4：Experimental Procedures and Operational Workflow:
The prosection method is applied to visualize the benchmark approximation sets and other approximation sets.
5：Data Analysis Methods:
The method's ability to preserve the Pareto dominance relation and relative closeness to reference points is analyzed theoretically.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
