The Garfield Demonstrator illustrates how Machine Learning is used to train a model on known data and how it fares on new data. It describes the Decentralized Learning model, showing how training on more data improves learning, while still keeping control on one’s data. Finally, it shows the negative impact that faulty components have on distributed learning, and how using Garfield helps to develop applications that can tolerate them.
The following screenshot shows the start of the demonstrator. A wizard-style interface will let you navigate through the various steps of the demonstrator.
For a Live Demo Click Here
Presentation
Demo
Technical