Federated Machine Learning with Kubernetes
summary
Shows a working federated training pipeline that uses TensorFlow Federated to train on an EMNIST-style dataset partitioned across 10 simulated clients, runs federated averaging for ~20 rounds with two optimizers, and reports decreasing loss and improving accuracy while never sharing raw examples. Simulation runs on a 3-node Kubernetes cluster with a deployment of TFF workers, a load balancer and gRPC orchestration, demonstrating reproducible client-server rounds and model aggregation.
AI-written summary · about
chapters (6)
- 0:00Start
- 0:51Machine Learning Systems
- 3:29Centralized Approach
- 4:12State of Machine Learning on the Edge
- 7:01Processing Data on the edge
- 8:59Federated Learning
related talks
31mFederated Machine Learning with Kubernetes
21mWhat, Why and How of Federated Machine Learning – Implementation Using...
26mDecentralized Federated Machine Learning: Empowering Edge Devices with Kubernetes
41mFederated Learning: Machine Learning on Decentralized Data
42mLearning Together, Distributed: An Introduction to Federated Learning
26mMachine Learning without the Data #federatedlearning
45mManaged federated learning on AWS: A case study for healthcare
40mDemystifying Decentralized Fair Models for Federated Machine Learning