Federated Learning Takes Flight Collaborative AI Training Soars

Federated Learning (FL) is a concept first introduced by Google in 2016, in which multiple devices collaboratively learn a machine learning model without sharing their private data under the supervision of a central server.

This offers ample opportunities in critical domains such as healthcare, finance etc, where it is risky to share private user information to other organisations or devices. While FL appears to be a promising Machine Learning (ML) technique to keep the local data private, it is also vulnerable to attacks like other ML models.

Artificial Intelligence (AI)/Machine Learning (ML) started getting popular in the last few 4-5 years, And one of the emerging trends in this field is Federated Learning.

Availability of Big-data and powerful computing units further accelerated the adoption of Machine Learning technologies in domains such as finance, healthcare, transportation, customer services, e-commerce, smart home applications etc. With this widespread adoption of ML techniques, it is therefore important to ensure the security and privacy of the techniques. In most of the machine learning applications, data from various organizations or devices are aggregated in a central server or a cloud platform for training the model. This is a key limitation especially when the training data set contains sensitive information and therefore, poses security threats.

 

TYPES OF FEDERATED LEARNING

Vertical Federated Learning -Vertical Federated Learning is used for cases in which each device contains dataset with different features but from sample instances. For instance, two organisations have data about the same group of people with different feature set can use Vertical FL to build a shared ML model.

Horizontal Federated Learning - Horizontal Federated Learning is used for cases in which each device contains dataset with the same feature space but with different sample instances. The first use case of FL- Google keyboard uses this type of learning in which the participating mobile phones have different training data with same features.

Federated Transfer Learning - Federated Transfer learning is similar to the traditional Machine Learning, where we want to add a new feature on a pre-trained model. The best example would be for giving an extension to the vertical federated learning - If we want to extend the ML to more number of sample instances which are not present in all of the collaborating organisations.

Cross-Silo Federated Learning - Cross- Silo Federated Learning is used when the participating devices are less in number and available for all rounds. The training data can be in horizontal or vertical FL format. Mostly cross-silo is used for cases with organisations. Works such as use cross-silo FL to develop their model.

Cross-Device Federated Learning - Scenarios with a large number of participating devices use Cross-device Federated Learning. Client-selection and incentive designs[37] are some notable techniques needed to facilitate this type of FL.

APPLICATIONS

Healthcare Electronic Health Records (EHR) is considered as the main source of healthcare data for machine learning applications . If ML models are trained only using the limited data available in a single hospital, it might introduce some amount of bias in the predictions.

Thus, to make the models more generalizable, it requires training with more data, which can be realized by sharing data among organizations. Given the sensitive nature of the healthcare data, it might not be feasible to share the electronic health records of patients among hospitals. In such situations, federated learning can serve as an option for building a collaborative learning model for healthcare data.

Transportation With the increase in the ubiquity of sensors in vehicular networks, it is feasible to capture more data and train ML models. Machine 2 Learning based models are generally applied to both vehicle management and traffic management . The current autonomous driving decisions are limited by the dynamic nature of the surroundings as the training is carried out offline.

FL can rescue such situations by online training vehicles from different geographical locations which can facilitate accurate labelling of the features. Similarly for traffic flow prediction techniques, a large amount of data is required, but most of the data is divided among various organizations and cannot be exchanged to protect the privacy . To address such situations also, we can deploy FL methods.

Finance One best use of federated learning in finance is in the banking sector, for loan risk assessment . Normally banks use whitelisting techniques to rule out the customers using their credit card reports from the central banks. Factors such as taxation, reputation etc can also be utilized for risk management by collaborating with other finance institutions and e-commerce companies. As it is risky to share private information of customers among organizations, they can make use of FL to build a risk assessment ML model.

Federated Learning offers a secure collaborative machine learning framework for different devices without sharing their private data. 4 This attracted a lot of researchers and there is extensive research happening in this domain. Federated Learning has been applied in several domains such as healthcare, transportation etc. Although FL frameworks offer a better privacy guarantee than other ML frameworks, it is still prone to several attacks. The distributed nature of the framework makes it even harder to deploy defense measures.