Interactive Evolutionary Multiobjective Optimization via Learning to Rank
Interactive Evolutionary Multiobjective Optimization via Learning to Rank
复制标题
基于排序学习的交互式进化多目标优化
DOI：
10.1109/tevc.2023.3234269
复制
发表时间：
2022-04
影响因子：
14.3
通讯作者：
Ke Li;Guiyu Lai;Xinghu Yao
中科院分区：
文献类型：
--
作者：
Ke Li;Guiyu Lai;Xinghu Yao
在实际的多目标决策中，如果决策者（DM）被要求在一组覆盖整个帕累托最优前沿的折衷方案中进行选择，这是很麻烦的。这是传统的进化多目标优化（EMO）的一个悖论，总是旨在实现收敛性和多样性之间的良好平衡。从本质上讲，多目标优化的最终目标是帮助DM识别感兴趣的解决方案（SOI），在多个冲突的标准之间实现令人满意的权衡。考虑到这一点，本文开发了一个框架，用于设计基于偏好的EMO算法，以交互方式找到SOI。其核心思想是将人纳入EMO的循环中。在每几次迭代之后，DM被邀请就几个现任候选人征求她的反馈。通过收集这些信息，她的偏好是逐步学习的学习排名神经网络，然后应用于指导基线EMO算法。注意，这个框架是如此通用，以至于任何现有的EMO算法都可以以插件的方式应用。48个基准测试问题，多达10个目标和现实世界的多目标机器人控制问题的实验充分证明了我们提出的算法寻找SOI的有效性。
In practical multicriterion decision making, it is cumbersome if a decision maker (DM) is asked to choose among a set of tradeoff alternatives covering the whole Pareto-optimal front. This is a paradox in conventional evolutionary multiobjective optimization (EMO) that always aim to achieve a well balance between convergence and diversity. In essence, the ultimate goal of multiobjective optimization is to help a DM identify solution(s) of interest (SOI) achieving satisfactory tradeoffs among multiple conflicting criteria. Bearing this in mind, this article develops a framework for designing preference-based EMO algorithms to find SOI in an interactive manner. Its core idea is to involve human in the loop of EMO. After every several iterations, the DM is invited to elicit her feedback with regard to a couple of incumbent candidates. By collecting such information, her preference is progressively learned by a learning-to-rank neural network and then applied to guide the baseline EMO algorithm. Note that this framework is so general that any existing EMO algorithm can be applied in a plug-in manner. Experiments on 48 benchmark test problems with up to ten objectives and a real-world multiobjective robot control problem fully demonstrate the effectiveness of our proposed algorithms for finding SOI.