内容提示: GAIN: Missing Data Imputation using Generative Adversarial NetsJinsung Yoon 1 * James Jordon 2 * Mihaela van der Schaar 1 2 3AbstractWe propose a novel method for imputing missingdata by adapting the well-known Generative Ad-versarial Nets (GAN) framework. Accordingly,we call our ...
ICML 2018 DOI:[1806.02920] GAIN: Missing Data Imputation using Generative Adversarial Nets Github:GitHub - jsyoon0823/GAIN: Generative Adversarial Imputation Networks (GAIN) Abstruct: 作者提出了基于GAN的数据补全方法。 Back Ground: 数据的缺失是一个很普遍的现象,有时因为数据本身就很难获得,有时是因为...
We propose a novel method for imputing missing data by adapting the well-known Generative Adversarial Nets (GAN) framework. Accordingly, we call our method Generative Adversarial Imputation Nets (GAIN). The generator (G) observes some components of a real data vector, imputes the missing component...
Generative Adversarial Imputation Networks (GAIN) Title: GAIN: Missing Data Imputation using Generative Adversarial Nets Authors: Jinsung Yoon, James Jordon, Mihaela van der Schaar Reference: J. Yoon, J. Jordon, M. van der Schaar, "GAIN: Missing Data Imputation using Generative Adversarial Nets,"...
of low-missing-rate data. Then an auxiliary classifier is determined using the syntheticpseudo-labels. Further, this classifier is incorporated into the generative adversarial framework to help the generator to yield higher quality imputation results. The proposed method can improve the imputation quality...
Paper: Jinsung Yoon, James Jordon, Mihaela van der Schaar, "GAIN: Missing Data Imputation using Generative Adversarial Nets," International Conference on Machine Learning (ICML), 2018. Paper Link:http://proceedings.mlr.press/v80/yoon18a/yoon18a.pdf ...
This paper presents three data imputation methods based on GAN: SGAIN, WSGAIN-CP and WSGAIN-GP. These methods were tested on datasets with different settings of missing values probabilities, where the values are missing completely at random (MCAR). The evaluation of the newly developed methods ...
This paper presents three data imputation methods based on GAN: SGAIN, WSGAIN-CP and WSGAIN-GP. These methods were tested on datasets with different settings of missing values probabilities, where the values are missing completely at random (MCAR). The evaluation of the newly developed methods ...
We compared the findings from our intent-to-treat analyses with those from (1) per-protocol models that included only data collected within the window (4 weeks after the 6-, 12-, and 18-month study visits) and (2) models using multiple imputation to replace missing and out-of-window dat...
For pre-pregnant weight and height, missing values were handled by multivariate imputation using chained equations (MICE) [30] based on the presence of GDM, and WGR before and after the OGTT. To test the robustness of our study, we conducted a sensitivity analysis by excluding those ...