REVIEW 11 cited by
Relational Deep Learning: Graph Representation Learning on Relational Databases
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Much of the world's most valued data is stored in relational databases and data warehouses, where the data is organized into many tables connected by primary-foreign key relations. However, building machine learning models using this data is both challenging and time consuming. The core problem is that no machine learning method is capable of learning on multiple tables interconnected by primary-foreign key relations. Current methods can only learn from a single table, so the data must first be manually joined and aggregated into a single training table, the process known as feature engineering. Feature engineering is slow, error prone and leads to suboptimal models. Here we introduce an end-to-end deep representation learning approach to directly learn on data laid out across multiple tables. We name our approach Relational Deep Learning (RDL). The core idea is to view relational databases as a temporal, heterogeneous graph, with a node for each row in each table, and edges specified by primary-foreign key links. Message Passing Graph Neural Networks can then automatically learn across the graph to extract representations that leverage all input data, without any manual feature engineering. Relational Deep Learning leads to more accurate models that can be built much faster. To facilitate research in this area, we develop RelBench, a set of benchmark datasets and an implementation of Relational Deep Learning. The data covers a wide spectrum, from discussions on Stack Exchange to book reviews on the Amazon Product Catalog. Overall, we define a new research area that generalizes graph machine learning and broadens its applicability to a wide set of AI use cases.
Forward citations
Cited by 11 Pith papers
-
Parameter-Free Encoders Remain Viable for RDB Foundation Models
Trainable RDB encoders cannot robustly exploit neighborhood labels as fixed foundation-model features or feature-importance signals, so simple parameter-free encoders stay near-SOTA.
-
UniSAGE: Unifying Static and Dynamic Attributes with Hyper-Structure
UniSAGE jointly models static and dynamic attributes in a single attribute graph and improves entity classification/regression over prior RDL and GNN baselines.
-
No Need to Train Your RDB Foundation Model
Column-wise, parameter-free JUICE encodings let single-table ICL models solve multi-table RDB prediction tasks with no training or fine-tuning.
-
Graph World Model
The Graph World Model uses action nodes and graph message passing to unify multimodal and graph-structured tasks, but its 'outperforms or matches' claim is contradicted by results on Goodreads.
-
Griffin: Towards a Graph-Centric Relational Database Foundation Model
A single pretrained graph model for relational databases, Griffin unifies encoders and decoders and reports the best average rank across 24 tasks in its evaluation.
-
A Self-Explainable Heterogeneous GNN for Relational Deep Learning
MPS-GNN learns predictive meta-paths in relational databases using aggregate statistics over their occurrences, not just existence, and outperforms prior heterogeneous GNNs in experiments.
-
Graph Neural Networks on Graph Databases
Training a GNN by querying a graph database for neighbor samples and features reduces memory use enough to train on small machines, but is much slower than in-memory training.
-
GraphAgent: Agentic Graph Language Assistant
GraphAgent links an LLM agent pipeline with graph tokenization, automatically building semantic knowledge graphs and using them for both graph prediction and open-ended text generation.
-
Tackling prediction tasks in relational databases with LLMs
Pre-trained LLMs, fed serialized relational rows with related examples, achieve competitive AUROC/MAE on RelBench without fine-tuning, but the headline comparison is weakened by pretraining contamination on Formula 1 tasks.
-
Graph-Based Feature Augmentation for Predictive Tasks on Relational Datasets
ReCoGNN automatically augments a base table with task-relevant features from related relational tables by splitting attributes into semantic sub-tables and propagating information through a weighted heterogeneous graph.
-
A Survey on Retrieval And Structuring Augmented Generation with Large Language Models
The paper presents a comprehensive survey and taxonomy of RAS methods, covering retrieval, text structuring, and LLM integration.
Discussion (0). Continue with ORCID to comment.