Scaling Blockchain ML With Databricks - From Graph Analytics to Graph Machine Learning
Databricks via YouTube
Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Discover how Coinbase leverages Databricks to scale machine learning on blockchain data in this 18-minute conference talk by Staff ML Engineer Indra Rustandi. Learn how to transform vast transaction networks into actionable insights using Databricks' scalable infrastructure powered by Delta Lake for real-time processing in ML applications like NFT floor price predictions. Explore the use of GraphFrames to analyze billion-node transaction graphs from networks like Bitcoin for clustering and fraud detection, uncovering structural patterns in blockchain data. Move beyond traditional graph analytics limitations by implementing Graph Neural Networks (GNNs) using Kumo AI, which learn directly from transaction networks rather than relying on hand-engineered features. Understand how GNNs encode relationships directly into models to adapt to new fraud tactics and capture subtle, evolving relationships over time. Gain insights into advancing blockchain machine learning through the combination of Databricks infrastructure and deep learning techniques applied to graph data structures.
Syllabus
Scaling Blockchain ML With Databricks: From Graph Analytics to Graph Machine Learning
Taught by
Databricks