Two Teachers Better Than One: Hardware-Physics Co-Guided Distributed Scientific Machine Learning

TL;DR

EPIC is a physics-guided distributed SciML framework that reduces communication costs and latency while maintaining or improving model accuracy, enabling real-time, energy-efficient in-field processing.

Contribution

It introduces a novel hardware- and physics-co-guided approach that combines local encoding and physics-aware decoding for distributed scientific machine learning.

Findings

Reduces latency by 8.9 times

Decreases communication energy by 33.8 times

Improves reconstruction fidelity on 8 out of 10 datasets

Abstract

Scientific machine learning (SciML) is increasingly applied to in-field processing, controlling, and monitoring; however, wide-area sensing, real-time demands, and strict energy and reliability constraints make centralized SciML implementation impractical. Most SciML models assume raw data aggregation at a central node, incurring prohibitively high communication latency and energy costs; yet, distributing models developed for general-purpose ML often breaks essential physical principles, resulting in degraded performance. To address these challenges, we introduce EPIC, a hardware- and physics-co-guided distributed SciML framework, using full-waveform inversion (FWI) as a representative task. EPIC performs lightweight local encoding on end devices and physics-aware decoding at a central node. By transmitting compact latent features rather than high-volume raw data and by using cross-attention to capture inter-receiver wavefield coupling, EPIC significantly reduces communication cost while preserving physical fidelity. Evaluated on a distributed testbed with five end devices and one central node, and across 10 datasets from OpenFWI, EPIC reduces latency by 8.9$\times$ and communication energy by 33.8$\times$, while even improving reconstruction fidelity on 8 out of 10 datasets.