Prospects of federated machine learning in fluid dynamics
This addresses the data centralization bottleneck for fluid dynamics researchers and practitioners, offering a decentralized alternative that could enhance privacy and efficiency, though it appears incremental as it applies an existing federated learning paradigm to this domain.
The paper tackles the problem of centralized data requirements in fluid dynamics machine learning by proposing a federated learning approach that allows localized clients to collaboratively train a shared model while keeping data on edge devices, demonstrating its feasibility for reconstructing spatiotemporal fields and suggesting it as a viable tool for accurate decentralized digital twins.
Physics-based models have been mainstream in fluid dynamics for developing predictive models. In recent years, machine learning has offered a renaissance to the fluid community due to the rapid developments in data science, processing units, neural network based technologies, and sensor adaptations. So far in many applications in fluid dynamics, machine learning approaches have been mostly focused on a standard process that requires centralizing the training data on a designated machine or in a data center. In this letter, we present a federated machine learning approach that enables localized clients to collaboratively learn an aggregated and shared predictive model while keeping all the training data on each edge device. We demonstrate the feasibility and prospects of such decentralized learning approach with an effort to forge a deep learning surrogate model for reconstructing spatiotemporal fields. Our results indicate that federated machine learning might be a viable tool for designing highly accurate predictive decentralized digital twins relevant to fluid dynamics.