| description abstract | Abstract. Foundation models are at the forefront of an increasing number of critical applications. In regards to technologies such as additive manufacturing (AM), these models have the potential to accelerate process design to build materials with optimized competing properties such as ductility and strength. However, lack of high-fidelity data is a major challenge that impedes the construction of such models for process–property optimization. To understand the impact of this challenge, and since foundation models rely on integrating or fusing multiple datasets, in this work, we conduct controlled experiments where we focus on the transferability of information across different material systems and properties (cross-material learning). More specifically, we generate experimental datasets from 17-4 PH and 316L stainless steels (SSs) in laser powder bed fusion (LPBF) where we measure the effect of five process parameters on two competing properties, namely, porosity and hardness. We then leverage various machine learning (ML) models such as Gaussian processes (GPs) and neural networks for process–property modeling in various configurations to test if knowledge about one material system or property can be leveraged to build more accurate ML models for process–property optimization. Through extensive cross-validation studies and probing the GPs’ interpretable hyperparameters, we study the intricate relation among data size and dimensionality, complexity of the process–property relations, noise, and characteristics of the ML models. Our findings highlight the need for structured learning approaches that incorporate domain knowledge in building foundation process–property models for design optimization rather than relying on uninformed data fusion in data-limited applications. | |
