Federated learning is a machine learning technique that can be used to train machine learning models at the location of data sources, by only communicating the trained models from individual data sources to reach a consensus for a global model. Therefore instead of using the traditional approach of collecting data from multiple sources to a centralized location for model training, this technique learns a collaborative model.

Federated learning addresses some of the major issues that prevail in the current machine learning technique such as data privacy, data security, data access rights, and access to data from heterogeneous sources.

There are different ways in which Federated Learning methods can be classified:

1. Based on the type of devices on which the ML models are trained on, there can be:

          a. Cross-Silo Federated Learning: ML models runs of the larger computing devices such as               cloud instances and bare metal servers.

          b. Cross-Device Federated Learning: ML models are run on edge or IoT devices such as               phones, or drones, or Raspberry Pi type of systems.

2. Based on data partitioning:

          a. Horizontal Federated Learning: The data coming from different sources have the same               feature space but they differ in data samples.

          b. Vertical Federated Learning: The data samples are aligned across the data sources but               they differ in data features.

          c. Federated Transfer Learning: In situations where neither the data samples or the feature               space are common between different data sources.

3. Based on communication methods:

          a. Centralized Federated Learning: a central server is used to orchestrate the different               steps of the algorithms and coordinate all the participating data sources during the               learning process.

          b. Decentralized Federated Learning: The data sources are able to coordinate themselves               to obtain the global model.

‍