Edge 275: Understanding Vertical Federated Learning

Vertical federated learning, Google's research about using federated learning to optimize mobile keyword predictions and the Flower framework.

Created Using Midjourney

In this issue:

• An overview of vertical federated learning

• A review of Google’s paper about using federated learning to optimize mobile keyword predictions.

• An introduction to Flower’s federated learning framework.

💡 ML Concept of the Day: Understanding Vertical Federated Learning

In the previous edition of this series, we explored horizontal federation learning(HFL) as an architecture in which nodes in a federation shared the same feature space but different sample space. The natural complement to HFL is a type of federation in which nodes share a similar sample space but different feature space. This is known in machine learning theory as vertical federated learning(VFL).