Physics-Informed Machine Learning Under Small-Data Constraints: Lessons from Abrasive Waterjet Milling
arXiv:2607.07863v1 Announce Type: new Abstract: In physically dominated machining processes, experimental datasets are small, expensive, and material-specific; in this regime, data curation, evaluation design, and the form of physics integration can matter as much as the learning algorithm. Using an abrasive waterjet milling dataset ($n{=}155$, Inconel\,718), we make three methodological contributions. First, we separate physics-based data \emph{cleaning} from statistical \emph{curation} and treat the latter as competing modelling hypotheses rather than silent preprocessing. Second, we find th
The increasing availability of powerful machine learning tools combined with the historical challenges of data collection in complex physical processes leads to innovations in physics-informed AI under small-data constraints.
This research provides a methodology for applying AI more effectively in industrial processes where large datasets are impractical, potentially accelerating innovation and efficiency in manufacturing and physical sciences.
The approach to integrating physics-based knowledge with statistical modeling for industrial AI applications, especially in contexts of limited experimental data, is refined and improved.
- · Industrial manufacturing companies
- · AI/ML researchers in physical sciences
- · Material science companies
- · Academic research institutions
- · Companies relying solely on large-data ML in physical domains
- · Sectors unwilling to integrate physics-based models
Improved efficiency and accuracy in manufacturing processes requiring precise material handling and shaping.
Reduced development costs and accelerated design cycles for new materials and complex components due to better predictive modeling.
Enhanced competitiveness for industries that can rapidly adapt and deploy physics-informed small-data AI, leading to new niches and market leaders.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG