Summary of Sampling-guided Heterogeneous Graph Neural Network with Temporal Smoothing For Scalable Longitudinal Data Imputation, by Zhaoyang Zhang et al.
Sampling-guided Heterogeneous Graph Neural Network with Temporal Smoothing for Scalable Longitudinal Data Imputation
by Zhaoyang Zhang, Ziqi Chen, Qiao Liu, Jinhan Xie, Hongtu Zhu
First submitted to arxiv on: 7 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel framework, Sampling-guided Heterogeneous Graph Neural Network (SHT-GNN), to tackle missing data imputation in longitudinal studies. Unlike traditional methods, SHT-GNN can handle arbitrary missing data patterns while maintaining computational efficiency. The approach models observations and covariates as distinct node types, connecting observation nodes at successive time points through subject-specific subnetworks. By leveraging mini-batch sampling and temporal smoothing, SHT-GNN efficiently scales to large datasets, learning node representations and imputing missing data effectively. Experiments on synthetic and real-world datasets, including the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset, demonstrate that SHT-GNN outperforms existing imputation methods even with high missing data rates. The results highlight SHT-GNN’s robust imputation capabilities and superior performance in complex longitudinal data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper proposes a new way to fill in gaps in long-term studies where some data is missing. Unlike other methods that require extra work to handle irregularly missing data, this approach can handle any type of missing data while still being efficient. The method uses graphs to connect the different types of data and samples specific groups of data to learn how to impute the missing parts. The results show that this new method is better than existing ones at filling in gaps even when a lot of data is missing. |
Keywords
» Artificial intelligence » Gnn » Graph neural network
