6.1 What is Taiwan AI Labs Federated Learning Framework?

Taiwan AI Labs Federated Learning Framework includes two major parts:

1. Edge Machines.

2. Aggregator.

The Edge Machine is installed inside Institutes, and doesn't need to have a public IP address to reduce the risk of disposing the machine in public domain. It could be just setup like other Institute's internal servers, and could only be reached by internal staffs.

The Aggregator needs to be set up with public IP (IPV4/V6) for communication with all Edge Machines. Trained model weights are aggregated there after each training epoch, and the final global model will be available there.

Note:

If you want to know the data flow of Taiwan AI Labs Federated Learning framework, please refer it here.

PS:

Below is a demonstration of Federated Learning progress, from creating the FL project, inviting participants, uploading datasets, initial AI model, weights, and then training the final result; by using Taiwan AI Labs Federated Learning Framework:

For more information of using Taiwan AI Labs Federated Learning Platform, please contact us directly.