Summary of Deep Generative Models For Subgraph Prediction, by Erfaneh Mahmoudzadeh et al.
Deep Generative Models for Subgraph Prediction
by Erfaneh Mahmoudzadeh, Parmis Naddaf, Kiarash Zahirnia, Oliver Schulte
First submitted to arxiv on: 7 Aug 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 This paper introduces a new task for deep graph learning called subgraph queries. Subgraph queries jointly predict components of a target subgraph based on evidence represented by an observed subgraph. For instance, a query can predict sets of target links and/or node labels. To answer these queries, the authors utilize a probabilistic Graph Generative Model, specifically a Variational Graph Auto-Encoder (VGAE) model augmented to represent joint distributions over links, node features, and labels. The authors use Bayesian optimization to tune the relative importance of these components in a specific domain. They describe deterministic and sampling-based inference methods for estimating subgraph probabilities from the generative graph distribution without retraining. These methods are applied on six benchmark datasets, achieving superior predictive performance with improvements in AUC scores ranging from 0.06 to 0.2 points. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Subgraph queries are a new way of using Graph Neural Networks (GNNs) that can model complex relational data. GNNs are useful for things like social network analysis and recommendation systems. Instead of just predicting individual components, subgraph queries try to predict the whole graph together. To do this, the authors use a special kind of AI model called a Variational Graph Auto-Encoder (VGAE). This model can generate new graphs that look similar to the ones it was trained on. The authors then use this model to make predictions about which links and nodes are in a target subgraph. They tested their method on six different datasets and found that it worked better than other methods. |
Keywords
» Artificial intelligence » Auc » Encoder » Generative model » Inference » Optimization
