Active Learning Guided Design Space Refinement for Scalable Multi-Objective Bayesian Optimization in Materials Discovery
Research paper by Alexandros Ntagiantas and collaborators. arXiv record dated 2026-08-05. Research area: Machine learning.
Resource information
- Authors
- Alexandros Ntagiantas, Panagiotis Tsilimidos, George Giannakopoulos, Christoforos Rekatsinas, Panagiotis Krokidas
- Research area
- Machine learning
- Application domain
- General AI research
- AI methods
- Machine learning
- Research topics
- cs.LG, cond-mat.mtrl-sci
- arXiv record date
- 2026-08-05
- Publication
- arXiv
- Upstream last modified
- Aug 6, 2026
- License
- CC BY 4.0
- Last record update
- Nov 4, 2026
Abstract
Advanced materials discovery increasingly relies on machine learning and Bayesian optimization to explore large discrete design spaces under limited evaluation budgets. However, conventional Bayesian optimization (BO) can become inefficient as candidate spaces grow, often evaluating low-value regions before reaching informative areas. We propose an active-learning (AL)-guided adaptive search-space refinement framework combined with multi-objective BO to accelerate materials optimization while preserving Pareto-relevant regions. We evaluate the approach on CH4/N2 separation in covalent-organic frameworks and pressure-vessel design with material-direction stress components and thickness objectives. Results show that the AL-guided refinement reduces the candidate space by approximately half while preserving more than 99 percent of the original hypervolume. The reduced-space strategy improves early convergence and cumulative Pareto-front discovery from the BO, demonstrating efficient large-scale materials optimization across constrained autonomous materials discovery settings.