Parallel Problem Solving from Nature, PPSN XI
Parallel Problem Solving from Nature, PPSN XI
复制标题
自然并行问题解决，PPSN XI
DOI：
10.1007/978-3-642-15844-5_39
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
Reynolds A
中科院分区：
文献类型：
--
作者：
Reynolds A
机器学习的目标是在未见过的数据上表现良好的预测模型。通常，这样的模型具有多种预期用途，与（例如）敏感性和特异性之间权衡的不同点相关。此外，当需要进行特征选择时，不同的特征子集将适合不同的目标性能特征。给定具有如此多个不同要求的特征选择任务，实际上面临着一个非常多目标的优化任务，其目标是特征子集的帕累托曲面，每个特征子集专门用于（例如）不同的灵敏度/特异性权衡配置文件。我们认为这种观点有很多优点。我们激励、开发和测试这种方法。我们证明，尽管目标数量任意多，但使用基于优势的多目标算法可以成功实现这一目标。
The target of machine learning is a predictive model that performs well on unseen data. Often, such a model has multiple intended uses, related to different points in the tradeoff between (e.g.) sensitivity and specificity. Moreover, when feature selection is required, different feature subsets will suit different target performance characteristics. Given a feature selection task with such multiple distinct requirements, one is in fact faced with a very-many-objective optimization task, whose target is a Pareto surface of feature subsets, each specialized for (e.g.) a different sensitivity/specificity tradeoff profile. We argue that this view has many advantages. We motivate, develop and test such an approach. We show that it can be achieved successfully using a dominance-based multiobjective algorithm, despite an arbitrarily large number of objectives.