Missing Pattern Recognized Diffusion Imputation Model for Missing Not At Random
Addresses the challenging MNAR imputation problem for practitioners in time-series and image domains, offering a principled framework that leverages missing patterns.
The paper proposes PRDIM, a diffusion-based imputation model that explicitly captures missing patterns under Missing Not at Random (MNAR) settings. It achieves strong imputation performance across time-series and image data, outperforming baselines in experiments.
Missing data frequently arises across diverse domains, including time-series and image domains. In the real world, missing occurrences often depend on the unobservable values themselves, which are referred to as Missing Not at Random (MNAR). In this work, we introduce the Missing Pattern Recognized Diffusion Imputation Model (PRDIM), a novel framework that explicitly captures the missing pattern and precisely imputes unobserved values. PRDIM iteratively maximizes the likelihood of the joint distribution for observed values and missing mask under an Expectation-Maximization (EM) algorithm. In this sense, we first employ a pattern recognizer, which approximates the underlying missing pattern and provides guidance during every inference toward more plausible imputations with respect to the missing information. Through extensive experiments, we demonstrate that PRDIM consistently achieves strong imputation performance under MNAR settings across multiple data modalities.