So far we have heard a lot about Convolutional Neural Networks (CNNs) which is a well-known method to handle euclidean data structures (like images, text, speech, and time series). However, in the real world, we are also surrounded by non-euclidean data like graphs, and a machine learning method to handle this type of data is known as the Graph Neural Networks. Therefore, in this workshop, we will deepen our knowledge with the concepts of Graph Neural Networks and compare them with CNNs.
This will be followed by an interactive Hands-On Session on Graph ML with a real-world application dataset. At the end, we will deploy our generated Graph ML model on the Triton inference server and perform Graph Analytics with the help of ArangoDB.
- Introduction to ArangoDB and Nvidia’s Triton Inference server (In Brief)
- Graph Machine Learning (GML) Motivation
- Applications of GML from various perspectives
- Get familiar with the working principles of GraphSage (Inductive Representation Learning algorithm)
- Train Graph ML (GraphSage) model using PyTorch Geometric library
- Dataset: Amazon co-purchasing dataset (OGB-Benchmark)
- Setting up Triton inference server on a local machine
- Deploy your first Graph ML model on Triton Server
- Write down python client-side script to interact with GML model served on Triton
- NVIDIA Triton meets ArangoDB: Use case for Amazon Product Recommendation
About the Presenter:
Sachin is a Machine Learning Research Engineer at ArangoDB whose aim is to build Intelligent products using thorough research and engineering in the area of Graph Machine Learning. He completed his Masters’s degree in Computer Science with a specialization in Intelligent Systems. He is an AI Enthusiast who has conducted research in the areas of Computer Vision, NLP, and Graph Neural Networks at DFKI (German Research Centre for AI) during his academic career. Sachin also worked on building Machine Learning pipelines at Define Media Gmbh where he worked as a Machine Learning Engineer and Scientist.