Flotta: a Secure and Flexible Spark-inspired Federated Learning Framework

TL;DR

Flotta is a Python-based federated learning framework inspired by Spark, enabling secure, flexible machine learning on sensitive distributed data within research consortia, especially in biomedical contexts.

Contribution

It introduces a novel Spark-inspired federated learning framework that emphasizes security, flexibility, and ease of use for sensitive multi-party research environments.

Findings

Demonstrates secure training on sensitive biomedical data

Highlights flexibility and user-friendliness of the framework

Provides a practical use case illustrating capabilities

Abstract

We present Flotta, a Federated Learning framework designed to train machine learning models on sensitive data distributed across a multi-party consortium conducting research in contexts requiring high levels of security, such as the biomedical field. Flotta is a Python package, inspired in several aspects by Apache Spark, which provides both flexibility and security and allows conducting research using solely machines internal to the consortium. In this paper, we describe the main components of the framework together with a practical use case to illustrate the framework's capabilities and highlight its security, flexibility and user-friendliness.