RefiDiff: Progressive Refinement Diffusion for Efficient Missing Data Imputation
This addresses the challenge of accurate and efficient data imputation for researchers and practitioners dealing with complex, high-dimensional datasets with missing values, particularly under difficult MNAR conditions.
The paper tackles the problem of missing data imputation in high-dimensional, mixed-type datasets under Missing Not At Random (MNAR) mechanisms by proposing RefiDiff, a framework that combines local predictions with a Mamba-based denoising network to capture long-range dependencies efficiently. It outperforms state-of-the-art methods across nine real-world datasets, showing strong performance in MNAR settings and superior out-of-sample generalization.
Missing values in high-dimensional, mixed-type datasets pose significant challenges for data imputation, particularly under Missing Not At Random (MNAR) mechanisms. Existing methods struggle to integrate local and global data characteristics, limiting performance in MNAR and high-dimensional settings. We propose an innovative framework, RefiDiff, combining local machine learning predictions with a novel Mamba-based denoising network efficiently capturing long-range dependencies among features and samples with low computational complexity. RefiDiff bridges the predictive and generative paradigms of imputation, leveraging pre-refinement for initial warm-up imputations and post-refinement to polish results, enhancing stability and accuracy. By encoding mixed-type data into unified tokens, RefiDiff enables robust imputation without architectural or hyperparameter tuning. RefiDiff outperforms state-of-the-art (SOTA) methods across missing-value settings, demonstrating strong performance in MNAR settings and superior out-of-sample generalization. Extensive evaluations on nine real-world datasets demonstrate its robustness, scalability, and effectiveness in handling complex missingness patterns.