Multi-Fidelity Adaptive Sampling

Multi-fidelity research focuses on combining models or data sources with varying levels of accuracy and computational cost to efficiently analyze and optimize complex systems. High-fidelity models offer precise insights but are computationally expensive, while low-fidelity models are faster but less accurate. By integrating these models, multi-fidelity approaches balance accuracy and efficiency, enabling better decision-making in resource-constrained environments.

Adaptive sampling in multi-fidelity research is crucial because it strategically selects the most informative data or model evaluations to maximize learning while minimizing computational effort. By intelligently exploring the design space, active learning enhances the synergy between high- and low-fidelity models, accelerating optimization and reducing the cost of developing robust solutions.

Journal Articles

Chen, Y. P., Wang, L., Comlek, Y., & Chen, W., “A Latent Variable Approach for Non-Hierarchical Multi-Fidelity Adaptive Sampling.” Computer Methods in Applied Mechanics and Engineering, 421 (2024), 116773.

Conference Talks

Chen, Y. P., Wang, L., Comlek, Y., & Chen, W., “Data Fusion of Multi-fidelity Systems via Latent Variable Gaussian Process for Active Learning Applications.” 2nd ICMA MMLDE-CSET, Sep. 23, 2023, El Paso, Texas, USA