Data is at the core of machine learning. Nevertheless, in many real-world projects, a single party’s data is often insufficient and needs to be augmented with data from other parties. However, there are also many concerns (regulatory, ethical, commercial etc.) stopping parties from exchanging data. Federated Learning is an emerging privacy-preserving machine learning technique. It enables multiple parties holding local data to collaboratively train machine learning models without exchanging their data with one another, hence preserving the confidentiality of different parties’ local data.
Synergos is an open-source Federated Learning platform which AI Singapore has been building. Synergos aims to make Federated Learning more accessible and sustainable.
Synergos makes Federated Learning Accessible
In conventional machine learning, it is commonly assumed that all data are from the same generative process and the generative process does not have memory of past generated data. This is something usually termed as IID (Independent and Identically Distributed) assumption. However, in Federated Learning, as different parties do not really see other parties’ data, it cannot be assumed that they all follow the same generative process. Non-IID data in Federated Learning could prevent the model from converging or take longer to converge. Many federated aggregation algorithms have been proposed to address this problem.
Synergos makes Federated Learning accessible by taking away the burden from the users in implementing those federated aggregation algorithms. The users could build a federated model by just declaring necessary information about what data they contribute and what model they want to build. Synergos coordinates different participating parties and does the heavy lifting to build the federated model.
Synergos makes Federated Learning sustainable
Usually, different parties incur non-negligible costs in acquiring and cleaning their data. They rarely altruistically share their data with others and risk losing their competitive edge. These parties would be more motivated to share their data when given enough incentives, Otherwise, without any party motivated to contribute data, which could be detrimental to the sustainability of Federated Learning.
Synergos makes Federated Learning sustainable by building a contribution & reward mechanism to reward different parties fairly based on their contributions.