Missing value imputation with adversarial random forests -- MissARF

2025-07-21

AI Summary

Missing data imputation in biostatistical analysis often faces a trade-off between accuracy and computational efficiency. This paper proposes MissARF, a generative imputation method based on Adversarial Random Forests (ARF), supporting both single and multiple imputation. Its core innovation lies in leveraging ARF to efficiently model high-dimensional conditional distributions and directly generating imputed values via conditional sampling; for multiple imputation, the same ARF model is reused without additional computational overhead. Experiments across diverse real-world and synthetic datasets demonstrate that MissARF achieves imputation accuracy comparable to state-of-the-art methods—including MICE and GAIN—while accelerating runtime by one to two orders of magnitude. The efficiency gain is especially pronounced for multiple imputation. Moreover, MissARF exhibits strong usability and scalability, making it suitable for large-scale biostatistical applications.

Technology Category

• Machine Learning: Ensemble Methods
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Handling missing values is a common challenge in biostatistical analyses, typically addressed by imputation methods. We propose a novel, fast, and easy-to-use imputation method called missing value imputation with adversarial random forests (MissARF), based on generative machine learning, that provides both single and multiple imputation. MissARF employs adversarial random forest (ARF) for density estimation and data synthesis. To impute a missing value of an observation, we condition on the non-missing values and sample from the estimated conditional distribution generated by ARF. Our experiments demonstrate that MissARF performs comparably to state-of-the-art single and multiple imputation methods in terms of imputation quality and fast runtime with no additional costs for multiple imputation.

Problem

• Develops MissARF for missing value imputation
• Uses adversarial random forests for density estimation
• Compares favorably to state-of-the-art imputation methods

Innovation

• Adversarial random forests for density estimation
• Conditional sampling from estimated distributions
• Fast runtime with no multiple imputation costs