Deployment-Aligned Low-Precision Neural Architecture Search for Spaceborne Edge AI

Parampuneet Kaur Thind, Vaibhav Kumar Katturu, Giacomo Zema, Roberto Del Prete

Keywords: Onboard processing, Edge AI, Low-precision training, On-device inference, Hardware-aware Neural Architecture Search, Maritime monitoring, Earth observation

TL;DR: We integrate deployment-aligned low-precision training into hardware-aware neural architecture search, improving on-device robustness of compact segmentation models on edge accelerators.

Abstract:

Designing deep networks that meet strict latency and accuracy constraints on edge accelerators increasingly relies on hardware-aware optimization, including neural architecture search (NAS) guided by device-level metrics. Yet most hardware-aware NAS pipelines still optimize architectures under full-precision assumptions and apply low-precision adaptation only after the search, leading to a mismatch between optimization-time behavior and deployment-time execution on low-precision hardware that can substantially degrade accuracy.
We address this limitation by integrating deployment-aligned low-precision training directly into the hardware-aware NAS loop. Candidate architectures are exposed to FP16 numerical constraints during fine-tuning and evaluation, enabling joint optimization of architectural efficiency and numerical robustness without modifying the search space or evolutionary strategy.
We evaluate the proposed framework on vessel segmentation for spaceborne maritime monitoring, targeting the Intel® Movidius Myriad X Vision Processing Unit (VPU) accelerator. While post-training precision conversion reduces on-device performance from 0.85 to 0.78 mIoU, deployment-aligned low-precision training achieves 0.826 mIoU on-device for the same architecture (95{,}791 parameters), recovering approximately two-thirds of the deployment-induced accuracy gap without increasing model complexity.
These results demonstrate that incorporating deployment-consistent numerical constraints into hardware-aware NAS substantially improves robustness and alignment between optimization and deployment for resource-constrained edge (Artificial Intelligence) AI systems.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 25