mlsanityData to 7 Oct 2026

FragmGAN: Generative Adversarial Nets for Fragmentary Data Imputation and Prediction

arXiv 2203.04692 ↗First posted 9 Mar 2022cs.LG cs.AIcited by 4

Not on the map. Why?

Abstract (arXiv)

Modern scientific research and applications very often encounter "fragmentary data" which brings big challenges to imputation and prediction. By leveraging the structure of response patterns, we propose a unified and flexible framework based on Generative Adversarial Nets (GAN) to deal with fragmentary data imputation and label prediction at the same time. Unlike most of the other generative model based imputation methods that either have no theoretical guarantee or only consider Missing Completed At Random (MCAR), the proposed FragmGAN has theoretical guarantees for imputation with data Missing At Random (MAR) while no hint mechanism is needed. FragmGAN trains a predictor with the generator and discriminator simultaneously. This linkage mechanism shows significant advantages for predictive performances in extensive experiments.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 19 of its 39 references that are in the map. First posted on arXiv 9 Mar 2022.

Builds on

19 of 39 references.

References · 19

  1. 2018GAIN: Missing Data Imputation using Generative Adversarial Netscited by 349
  2. 2020Handling Missing Data with Graph Representation Learningcited by 132
  3. 2019MisGAN: Learning from Incomplete Data with Generative Adversarial Networkscited by 74
  4. 2018Variational Autoencoder with Arbitrary Conditioningcited by 68
  5. 2020not-MIWAE: Deep Generative Modelling with Missing not at Random Datacited by 61
  6. 2020MCFlow: Monte Carlo Flow Models for Data Imputationcited by 37
  7. 2019CollaGAN : Collaborative GAN for Missing Image Data Imputationcited by 33
  8. 2019Missing Not at Random in Matrix Completion: The Effectiveness of Estimating Missingness Probabilities Under a Low Nuclear Norm Assumptioncited by 29
  9. 2017MIDA: Multiple Imputation using Denoising Autoencoderscited by 20
  10. 2019Improving Missing Data Imputation with Deep Generative Modelscited by 19
  11. 2020PC-GAIN: Pseudo-label Conditional Generative Adversarial Imputation Networks for Incomplete Datacited by 15
  12. 2019Integrating multi-source block-wise missing data in model selectioncited by 11
7 more
  1. 2019Integrative Factor Regression and Its Inference for Multimodal Data Analysiscited by 11
  2. 2019HexaGAN: Generative Adversarial Nets for Real World Classificationcited by 10
  3. 2021Variational Selective Autoencoder: Learning from Partially-Observed Heterogeneous Datacited by 10
  4. 2020Missing Features Reconstruction Using a Wasserstein Generative Adversarial Imputation Networkcited by 5
  5. 2020Imputation of Missing Data with Class Imbalance using Conditional Generative Adversarial Networkscited by 4
  6. 2019Unsupervised Data Imputation via Variational Inference of Deep Subspacescited by 2
  7. 2020IFGAN: Missing Value Imputation using Feature-specific Generative Adversarial Networkscited by 1

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 2.

4 papers.

  1. 2023Transformed Distribution Matching for Missing Value Imputationcited by 31
  2. 2024Optimal Transport for Structure Learning Under Missing Datacited by 5
  3. 2024A Context-Aware Approach for Enhancing Data Imputation with Pre-trained Language Modelscited by 5
  4. 2023ClueGAIN: Application of Transfer Learning On Generative Adversarial Imputation Nets (GAIN)cited by 0