MIDA: Multiple Imputation using Denoising Autoencoders

Abstract

Missing data is a significant problem impacting all domains. State-of-the-art framework for minimizing missing data bias is multiple imputation, for which the choice of an imputation model remains nontrivial. We propose a multiple imputation model based on overcomplete deep denoising autoencoders. Our proposed model is capable of handling different data types, missingness patterns, missingness proportions and distributions. Evaluation on several real life datasets show our proposed model significantly outperforms current state-of-the-art methods under varying conditions while simultaneously improving end of the line analytics.

To appear in the proceedings of the 22nd Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2018)

References in corpus (1)

• Missing Data: A Comparison of Neural Network and Expectation Maximisation Techniques

Cited by in corpus (7)

• GAIN: Missing Data Imputation using Generative Adversarial Nets
• Stacked DeBERT: All Attention in Incomplete Data for Text Classification
• Improving Missing Data Imputation with Deep Generative Models
• Learning representations for multivariate time series with missing data using Temporal Kernelized Autoencoders
• Blood lactate concentration prediction in critical care patients: handling missing values
• PC-GAIN: Pseudo-label Conditional Generative Adversarial Imputation Networks for Incomplete Data
• DPER: Efficient Parameter Estimation for Randomly Missing Data