The Problem Solving Benefits of Down-sampling Vary by Selection Scheme
The Problem Solving Benefits of Down-sampling Vary by Selection Scheme
复制标题
下采样解决问题的好处因选择方案而异
DOI：
10.1145/3583133.3590713
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Lalejini, Alexander
中科院分区：
文献类型：
--
作者：
Boldi, Ryan;Bao, Ashley;Briesch, Martin;Helmuth, Thomas;Sobania, Dominik;Spector, Lee;Lalejini, Alexander
遗传编程系统通常使用大型训练集来评估候选解决方案，这可能是计算昂贵的。下采样训练集长期以来一直被用于降低评估的计算成本，在广泛的应用领域。事实上，最近的研究表明，随机和知情的下采样都可以大大提高使用词典亲本选择的GP系统解决问题的成功率。我们使用PushGP框架进行实验测试，这些下采样技术是否也可以提高解决问题的成功率，在其他两个常用的选择方法，健身比例和锦标赛选择的背景下，在八个GP问题（四个程序合成和四个符号回归）。我们验证了降采样可以帮助适应度比例和锦标赛选择解决问题的成功。然而，问题的数量，其中下采样改善解决问题的成功不同的选择方案，这表明下采样的影响取决于问题和选择方案。令人惊讶的是，我们发现，与锦标赛和适应度比例选择相比，下采样与词库选择相结合时最一致地有益。总的来说，我们的研究结果表明，下采样时，应考虑更经常解决基于测试的GP问题。
Genetic programming systems often use large training sets to evaluate candidate solutions, which can be computationally expensive. Down-sampling training sets has long been used to decrease the computational cost of evaluation in a wide range of application domains. Indeed, recent studies have shown that both random and informed down-sampling can substantially improve problem-solving success for GP systems that use lexicase parent selection. We use the PushGP framework to experimentally test whether these down-sampling techniques can also improve problem-solving success in the context of two other commonly used selection methods, fitness-proportionate and tournament selection, across eight GP problems (four program synthesis and four symbolic regression). We verified that down-sampling can benefit the problem-solving success of both fitness-proportionate and tournament selection. However, the number of problems wherein down-sampling improved problem-solving success varied by selection scheme, suggesting that the impact of down-sampling depends both on the problem and choice of selection scheme. Surprisingly, we found that down-sampling was most consistently beneficial when combined with lexicase selection as compared to tournament and fitness-proportionate selection. Overall, our results suggest that down-sampling should be considered more often when solving test-based GP problems.
登录
查看更多内容
知情的下采样 Lexicase 选择：识别有效的培训案例以高效解决问题
影响因子：
6.8
作者：
Ryan Boldi;Martin Briesch;Dominik Sobania;Alexander Lalejini;Thomas Helmuth;Franz Rothlauf;Charles Ofria;L. Spector
通讯作者：
L. Spector
随机子采样提高了词典选择的性能
DOI：
10.1145/3319619.3326900
发表时间：
2019
期刊：
GECCO '19: Proceedings of the Genetic and Evolutionary Computation Conference
影响因子：
--
作者：
Hernandez, Jose Guadalupe;Lalejini, Alexander;Dolson, Emily;Ofria, Charles
通讯作者：
Ofria, Charles
影响因子：
2.6
作者：
L. Spector;Alan J. Robinson
通讯作者：
L. Spector;Alan J. Robinson
探索中的探索：衡量词汇选择的能力，以找到通往最优性的模糊途径
DOI：
10.1007/978-981-16-8113-4_5
发表时间：
2021
期刊：
ArXiv
影响因子：
--
作者：
Jose Guadalupe Hernandez;Alexander Lalejini;Charles Ofria
通讯作者：
Charles Ofria
关于进化算法中有用的多样性，系统发育指标可以告诉我们什么？
DOI：
10.1007/978-981-16-8113-4_4
发表时间：
2021
期刊：
ArXiv
影响因子：
--
作者：
Jose Guadalupe Hernandez;Alexander Lalejini;Emily L. Dolson
通讯作者：
Emily L. Dolson