Ensemble Generative Adversarial Imputation Network with Selective Multi-Generator (ESM-GAIN) for Missing Data Imputation
Ensemble Generative Adversarial Imputation Network with Selective Multi-Generator (ESM-GAIN) for Missing Data Imputation
复制标题
DOI：
10.1109/case49997.2022.9926629
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
Yuxuan Li;Ayse Dogan;Chenang Liu
中科院分区：
文献类型：
--
作者：
Yuxuan Li;Ayse Dogan;Chenang Liu
作为一个普遍存在的问题，数据缺失可能会影响数据建模的性能，并导致完成预期任务的更多困难。已经开发了许多用于缺失数据补偿的方法。最近，利用新兴的生成性对抗性网络(GAN)，提出了一种有效的缺失数据填补方法--生成性对抗性填补网(Gain)。然而，它的建模体系结构仍然可能导致显著的归罪偏差。此外，对于GaN结构，增益的训练过程可能是不稳定的，并且归属变化可能很大。因此，为了克服这两个局限性，提出了选择多生成器集成增益(ESM-Gain)来提高估计精度和稳健性。提出的ESM-Gain的贡献包括两个方面：(1)提出了一种选择性多代框架来识别高质量的推算；(2)加入了集成学习框架来进行增益推算，以提高推算的稳健性。数值模拟和两个真实乳腺癌数据集验证了所提出的ESM-Gain算法的有效性。
As a pervasive issue, missing data may influence the data modeling performance and lead to more difficulties of completing the desired tasks. Many approaches have been developed for missing data imputation. Recently, by taking advantage of the emerging generative adversarial network (GAN), an effective missing data imputation approach termed generative adversarial imputation nets (GAIN) was developed. However, its modeling architecture may still lead to significant imputation bias. In addition, with the GAN structure, the training process of GAIN may be instable and the imputation variation may be high. Hence, to address these two limitations, the ensemble GAIN with selective multi-generator (ESM-GAIN) is proposed to improve the imputation accuracy and robustness. The contributions of the proposed ESM-GAIN consist of two aspects: (1) a selective multi-generation framework is proposed to identify high-quality imputations; (2) an ensemble learning framework is incorporated for GAIN imputation to improve the imputation robustness. The effectiveness of the proposed ESM-GAIN is validated by both numerical simulation and two real-world breast cancer datasets.