Runtime-Robust Edge Inference System with Masking-Based Partial Update on Dynamic Reconfigurable FPGA

Myeongjin Kang, Daejin Park · 2025

Read Runtime-Robust Edge Inference System with Masking-Based Partial Update on Dynamic Reconfigurable FPGA by Myeongjin Kang, Daejin Park (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12736569.txt.

Description

Myeongjin Kang and Daejin Park build an edge AI system that updates itself in place in a 2025 paper. Devices at the network edge run vision models under changing light, noise, and new object types. A frozen model loses accuracy as conditions drift. Full retraining from the cloud is slow and costly in data traffic.

Their design splits device memory into an active zone and an adaptive zone. A server-side programmable chip judges which layers matter most, retrains just those parts, and sends a small masked update. The device swaps the update in without pausing. On three image data sets, adaptation ran up to 1.3 times faster than full GPU retraining. Update traffic fell to 28 percent of a full model transfer.

Catalog record

CREATORMyeongjin Kang, Daejin Park
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE31 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC12736569

Browse the full library · About the Tower of Babel