NASPrecision: Neural Architecture Search-Driven Multi-Stage Learning for surface roughness prediction in ultra-precision machining

Penghui Ruan, Divya Saxena, Jiannong Cao, Xiao-Yun Liu, Ruoxin Wang, Chi Fai Cheung

Abstract: Highlights•Propose a neural architecture search driven framework for ultra precision machining.•Propose a multi-stage learning process to correct bias.•Incorporate generative models to enrich and balance the datasets.•Achieve better performance on three distinct datasets compared to existing methods.