Extended Missing Data Imputation via GANs for Ranking Applications

TL;DR

This paper introduces Conditional Imputation GAN, a novel method for missing data imputation tailored for ranking applications, which relaxes traditional assumptions and improves imputation quality, especially in ranking datasets.

Contribution

The paper presents a GAN-based imputation method that handles complex missing data mechanisms in ranking datasets, extending beyond standard assumptions and providing theoretical guarantees.

Findings

Achieves superior imputation quality on MSR and synthetic datasets.

Demonstrates comparable ranking performance using GAN-imputed data on Amazon Search.

Proves optimal GAN imputation under EMAR and EAMAR mechanisms.

Abstract

We propose Conditional Imputation GAN, an extended missing data imputation method based on Generative Adversarial Networks (GANs). The motivating use case is learning-to-rank, the cornerstone of modern search, recommendation system, and information retrieval applications. Empirical ranking datasets do not always follow standard Gaussian distributions or Missing Completely At Random (MCAR) mechanism, which are standard assumptions of classic missing data imputation methods. Our methodology provides a simple solution that offers compatible imputation guarantees while relaxing assumptions for missing mechanisms and sidesteps approximating intractable distributions to improve imputation quality. We prove that the optimal GAN imputation is achieved for Extended Missing At Random (EMAR) and Extended Always Missing At Random (EAMAR) mechanisms, beyond the naive MCAR. Our method demonstrates the highest imputation quality on the open-source Microsoft Research Ranking (MSR) Dataset and a synthetic ranking dataset compared to state-of-the-art benchmarks and across various feature distributions. Using a proprietary Amazon Search ranking dataset, we also demonstrate comparable ranking quality metrics for ranking models trained on GAN-imputed data compared to ground-truth data.