Decentralized Federated Machine Learning: Empowering Edge Devices with Kubernetes

Explore a 26-minute conference talk that delves into the implementation of Federated Machine Learning on edge devices using Kubernetes orchestration. Learn how to overcome the challenges of enhancing machine learning models while maintaining data privacy through decentralized training approaches. Discover the fundamental differences between traditional and federated learning methods, and understand how Kubernetes enables scalable orchestration of federated learning computations. Through a practical image classification demonstration, examine the benefits of this approach including reduced latency, improved power efficiency, and enhanced privacy protection. Master the techniques for revolutionizing model development using privacy-preserving, Kubernetes-powered Federated Machine Learning, presented by experts from NYU and SigNoz at the Cloud Native Computing Foundation event.