Table of Contents

1. Introduction
2. Federated Learning on Mobile with NVIDIA FLARE and Meta ExecuTorch
3. Collaborative Model Learning on Distributed Edge Devices
4. Effortless Cross-Device Federated Pipeline Development

Introduction

• Effortless development: Abstracted device complexity, handling hardware, OS, ML frameworks, and programming languages for seamless FL on mobile.
• Collaborative model learning on distributed edge devices

Federated Learning on Mobile with NVIDIA FLARE and Meta ExecuTorch

• FL environment capable of orchestrating learning flow with millions of devices.
• NVIDIA FLARE and ExecuTorch integration for cross-device FL paradigm.
• Hierarchical FL architecture for efficient management of edge devices.
• Reliable and scalable model training across distributed mobile devices while preserving data privacy.

Collaborative Model Learning on Distributed Edge Devices

• Many AI models rely on data generated at edge devices in daily life.
• Challenges in training robust AI models based on data collected at edge devices.
• High scalability of participating devices and web gateways for device connectivity.

Effortless Cross-Device Federated Pipeline Development

• Streamlined cross-device FL pipeline for efficient workflow.
• Customizable components for focusing on innovation without infrastructure concerns.