Sample-Efficient Bayesian Transfer Learning for Online Machine Parameter Optimization

Philipp Wagner 1 , Tobias Nagel 1 , Philipp Leube 2 and Marco F. Huber 1,3 · 2025

Read Sample-Efficient Bayesian Transfer Learning for Online Machine Parameter Optimization by Philipp Wagner 1 , Tobias Nagel 1 , Philipp Leube 2 and Marco F. Huber 1,3 (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/arxiv-2503-15928.txt.

Description

Philipp Wagner and co-authors tune factory machines with borrowed data in a 2025 paper. Setting a production machine well lifts quality and cuts waste, but each trial run costs time and stock. Their method runs a Bayesian search on the machine itself, which picks each next trial with care.

Past data from like machines seed the search through transfer learning, so good settings show up in few tries. The team proved the method on a real laser cutter for sheet metal in a working plant. Fewer failed cuts mean lower cost and less scrap, which backs both profit and green goals. The work shows on site tuning can pay off fast.

Catalog record

CREATORPhilipp Wagner 1 , Tobias Nagel 1 , Philipp Leube 2 and Marco F. Huber 1,3
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE31 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-15928

Browse the full library · About the Tower of Babel