Active Learning for Continual Learning & Sequential Decision-Making

Sequential decision-making is the key to active data collection for training data-driven surrogate models and to design optimization for simulation-based design. By leveraging quantified model uncertainty, I design algorithms that enable different types of active learning to identify the queries that maximize the improvement of the system, thereby maximizing sampling efficiency in complex system design and Digital Twins.

This stream of research contributes to various engineering domains, including data fusion and design optimization using multi-fidelity data sources and integrating simulation and experimental data, and connects to autonomous discovery in materials design. Future directions include leveraging agentic systems to connect active-learning algorithms with experimental and simulation platforms for self-discovery.