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," International Conference on Machine Learning (ICML), 2018.

Paper Link: http://medianetlab.ee.ucla.edu/papers/ICML_GAIN.pdf

Appendix Link: http://medianetlab.ee.ucla.edu/papers/ICML_GAIN_Supp.pdf

Description of the code

This code shows the implementation of GAIN on MNIST dataset.

1. Introducing 50% of missingness on MNIST dataset.

2. Recover missing values on MNIST datasets using GAIN.

3. Show the multiple imputation results on MNIST with GAIN.

---

Add source codes for UCI Letter and Spam datasets (02/12/2019)