mlsanityData to 14 Oct 2026

Simultaneous Missing Value Imputation and Structure Learning with Groups

arXiv 2110.08223 ↗First posted 15 Oct 2021cs.LGcited by 16

Not on the map. Why?

Abstract (arXiv)

Learning structures between groups of variables from data with missing values is an important task in the real world, yet difficult to solve. One typical scenario is discovering the structure among topics in the education domain to identify learning pathways. Here, the observations are student performances for questions under each topic which contain missing values. However, most existing methods focus on learning structures between a few individual variables from the complete data. In this work, we propose VISL, a novel scalable structure learning approach that can simultaneously infer structures between groups of variables under missing data and perform missing value imputations with deep learning. Particularly, we propose a generative model with a structured latent space and a graph neural network-based architecture, scaling to a large number of variables. Empirically, we conduct extensive experiments on synthetic, semi-synthetic, and real-world education data sets. We show improved performances on both imputation and structure learning accuracy compared to popular and recent approaches.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). Builds on 31 of its 75 references that are in the map. First posted on arXiv 15 Oct 2021.

Builds on

31 of 75 references.

References · 31

  1. 2013Auto-Encoding Variational Bayescited by 4,053
  2. 2017Neural Message Passing for Quantum Chemistrycited by 3,487
  3. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  4. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
  5. 2019An Introduction to Variational Autoencoderscited by 1,000
  6. 2018DAGs with NO TEARS: Continuous Optimization for Structure Learningcited by 752
  7. 2019DAG-GNN: DAG Structure Learning with Graph Neural Networkscited by 331
  8. 2019Gradient-Based Neural DAG Learningcited by 227
  9. 2019Learning Sparse Nonparametric DAGscited by 207
  10. 2020On the Role of Sparsity and DAG Constraints for Learning Linear DAGscited by 191
  11. 2018Handling Incomplete Heterogeneous Data using VAEscited by 149
  12. 2021SLAPS: Self-Supervision Improves Structure Learning for Graph Neural Networkscited by 110
19 more
  1. 2021DiBS: Differentiable Bayesian Structure Learningcited by 106
  2. 2020Implicit Graph Neural Networkscited by 100
  3. 2013High-dimensional learning of linear causal networks via inverse covariance estimationcited by 87
  4. 2018EDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAEcited by 87
  5. 2018Variational Autoencoder with Arbitrary Conditioningcited by 68
  6. 2018Causal Discovery in the Presence of Missing Datacited by 47
  7. 2021Unsuitability of NOTEARS for Causal Graph Discoverycited by 47
  8. 2018Knowledge Tracing Machines: Factorization Machines for Knowledge Tracingcited by 42
  9. 2020CASTLE: Regularization via Auxiliary Causal Graph Discoverycited by 41
  10. 2020Towards Scalable Bayesian Learning of Causal DAGscited by 33
  11. 2019Neuropathic Pain Diagnosis Simulator for Causal Discovery Algorithm Evaluationcited by 32
  12. 2014Concave Penalized Estimation of Sparse Gaussian Bayesian Networkscited by 28
  13. 2017Fast Causal Inference with Non-Random Missingness by Test-Wise Deletioncited by 23
  14. 2021Results and Insights from Diagnostic Questions: The NeurIPS 2020 Education Challengecited by 18
  15. 2018Zero Shot Learning for Code Education: Rubric Sampling with Deep Learning Inferencecited by 16
  16. 2014Penalized Estimation of Directed Acyclic Graphs From Discrete Datacited by 10
  17. 2013Time-varying Learning and Content Analytics via Sparse Factor Analysiscited by 6
  18. 2018Conditional Inference in Pre-trained Variational Autoencoders via Cross-codingcited by 4
  19. 2019Icebreaker: Element-wise Active Information Acquisition with Bayesian Deep Latent Gaussian Modelcited by 3

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 8; 2024: 5; 2025: 2.

16 papers.

Sort:

By area:

  1. 2023CUTS+: High-dimensional Causal Discovery from Irregular Time-seriescited by 41
  2. 2023CUTS: Neural Causal Discovery from Irregular Time-Series Datacited by 33
  3. 2023Transformed Distribution Matching for Missing Value Imputationcited by 31
  4. 2023Causal Deep Learningcited by 20
  5. 2023Causal Reasoning in the Presence of Latent Confounders via Neural ADMG Learningcited by 15
  6. 2023RedCore: Relative Advantage Aware Cross-modal Representation Learning for Missing Modalities with Imbalanced Missing Ratescited by 15
  7. 2024Causality for Tabular Data Synthesis: A High-Order Structure Causal Benchmark Frameworkcited by 8
  8. 2023Predictive Coding beyond Correlationscited by 6
  9. 2024Optimal Transport for Structure Learning Under Missing Datacited by 6
  10. 2024A Context-Aware Approach for Enhancing Data Imputation with Pre-trained Language Modelscited by 5
  11. 2023Knowledge Guided Representation Learning and Causal Structure Learning in Soil Sciencecited by 3
  12. 2024Data Augmentation for Sparse Multidimensional Learning Performance Data Using Generative AIcited by 3
4 more
  1. 2025Incomplete Graph Learning: A Comprehensive Surveycited by 3
  2. 2024Generative Adversarial Networks for Imputing Sparse Learning Performancecited by 2
  3. 2022GLACIAL: Granger and Learning-based Causality Analysis for Longitudinal Imaging Studiescited by 1
  4. 2025Generative Data Imputation for Sparse Learner Performance Data Using Generative Adversarial Imputation Networkscited by 0