Plankton-FL: Exploration of Federated Learning for Privacy-Preserving Training of Deep Neural Networks for Phytoplankton Classification

TL;DR

This paper investigates federated learning as a privacy-preserving method for training deep neural networks to classify phytoplankton, comparing it to centralized learning through simulation experiments.

Contribution

It explores the feasibility of federated learning frameworks for phytoplankton classification and compares their performance to traditional centralized approaches.

Findings

Federated learning is feasible for phytoplankton monitoring.

FL and ME-FL outperform centralized learning in privacy preservation.

Experimental results show comparable accuracy to centralized methods.

Abstract

Creating high-performance generalizable deep neural networks for phytoplankton monitoring requires utilizing large-scale data coming from diverse global water sources. A major challenge to training such networks lies in data privacy, where data collected at different facilities are often restricted from being transferred to a centralized location. A promising approach to overcome this challenge is federated learning, where training is done at site level on local data, and only the model parameters are exchanged over the network to generate a global model. In this study, we explore the feasibility of leveraging federated learning for privacy-preserving training of deep neural networks for phytoplankton classification. More specifically, we simulate two different federated learning frameworks, federated learning (FL) and mutually exclusive FL (ME-FL), and compare their performance to a traditional centralized learning (CL) framework. Experimental results from this study demonstrate the feasibility and potential of federated learning for phytoplankton monitoring.