TigerGraph ML Workbench
TigerGraph Machine Learning Workbench is a platform designed for data scientists and AI/ML practitioners to easily develop graph-based machine learning models with production-scale graph data stored in TigerGraph.
It is available as a service through TigerGraph Cloud or standalone in two editions, Developer and Enterprise. It provides robust and efficient data pipelines at the Python level to interact with the TigerGraph database to perform common data processing functions such as:
graph feature generation
data batching and/or streaming for ML model development, training, and inference purposes.
ML Workbench on Cloud provides an ML solution from development to deployment including a fully-managed Jupyter notebook instance with:
distributed AutoML for hyperparameter tuning
ML pipeline management & orchestration
model deployment and serverless production serving
ML Workbench also comes in standalone Developer or Enterprise editions as a Jupyter-based Python development framework for direct integration with existing machine learning infrastructure.
It is compatible with other popular ML frameworks such as PyTorch Geometric, DGL, and TensorFlow as well as cloud infrastructure including Amazon SageMaker, Google Vertex AI and Microsoft Azure.
GNNs tend to outperform other machine learning techniques when there are well-defined relationships between data as it directly models the connectivity of your graph data. In recent research, GNNs have proven their success across various business domains and applications. With TigerGraph ML Workbench, you can now easily explore the potentials of GNN for your domains.
Here are some papers and resources to spark ideas in a range of applications and industries:
- Recommendation Engines
Pinterest introduced PinSAGE, an architecture that can serve real-time recommendations to their users, resulting in a 10-30% improvement compared to other deep learning methods when evaluated in A/B testing.
- Supply Chain
Amazon released a GNN architecture that incorporates temporal information with GNNs for demand forecasting. The method models interactions between products and their sellers on Amazon in a graph, resulting in a 16% improvement over other state-of-the-art forecasting methods.
AstraZeneca has used graph neural networks like GraphSAGE to generate knowledge graph embeddings for predicting possible drug-drug interactions such as possible synergies between drugs, as well as possible polypharmacy side effects. Additionally, the possibility of repurposing drugs to treat COVID has been studied using a drug repurposing knowledge graph and GNNs.
- Financial Institutions
GCNs have been studied for predicting money-laundering behavior in Bitcoin transaction networks, and have been shown to perform admirably compared to other approaches.
If you are interested in learning more about the fundamental research on different variations of GNNs, here is a list of helpful publications: