Learned Execution Graphs for Anomaly Detection & Drift in APIs — Ritvik Pandya, JP Morgan Chase
This talk introduces Learned Execution Graphs (LEGs) as a novel approach to detecting anomalies and drift in API usage. The core thesis is that by modeling API interactions as graphs and learning their execution patterns, systems can proactively identify deviations from normal behavior, which is crucial for maintaining API reliability and security.
20 min