Interpretable Meta-Learning for Multi-Objective Chemical Search
Abstract
Navigating the vast space of synthetically accessible molecules demands surrogate models that are interpretable and capable of handling multiple competing objectives at the same time. Deep learning approaches struggle to satisfy them under the computational constraints of quantum-level chemistry. Here, we introduce a modular pipeline that combines interpretable linear meta-learning models and adaptive-confidence uncertainty quantification into an Efficient Global Optimization (EGO) framework for multi-objective molecular discovery. For the first time, linear meta-learning is deployed in a multi-objective chemical search setting: by training across chemical objectives and cheap auxiliary properties, the meta-learned surrogates acquire transferable chemical knowledge that adapts rapidly to new objectives from limited data. Evaluated empirically on a real large scale search for spin-crossover metal-organic complexes, the baseline performs 78% worse in Pareto sense than the meta-learning alternative. We also address the calibration challenges inherent to active search. Since optimal candidates typically lie precisely in the distributional tails, standard uncertainty estimates fail. We introduce an adaptive confidence-tuning algorithm that dynamically recalibrates the exploration-exploitation trade-off as the molecular search evolves. Empirically, dynamic confidence tuning further dominates over 50% of the statically calibrated front.
Related
- Query-Efficient Quantum Multi-Objective Optimization for Materials Configuration Search via Bayesian Weight Selection
- Evolution of fairness in multi-objective reinforcement learning framework
- Physics-guided inverse design of Co-based superalloys using machine learning and multi-objective optimization for enhanced γ' solvus temperature
- VESPER: Online Learning of Free Energies with a Multi-Tempered Variational Objective
- CFD and machine-learning-assisted multi-objective optimization of trailing-edge serrations and vortex generators for flow-separation control on a NACA 0012 Airfoil
BibTeX
@misc{varagnolo2026interpretable,
title = {{Interpretable Meta-Learning for Multi-Objective Chemical Search}},
author = {Antonio Varagnolo and Yulia Pimonova and Michael G. Taylor and Raphaël Pestourie and Nicholas E. Lubbers},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2606.20497},
eprint = {2606.20497},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.20497},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Interpretable Meta-Learning for Multi-Objective Chemical Search
AU - Antonio Varagnolo
AU - Yulia Pimonova
AU - Michael G. Taylor
AU - Raphaël Pestourie
AU - Nicholas E. Lubbers
PY - 2026
DA - 2026/10/20
PB - arXiv
DO - 10.48550/arxiv.2606.20497
UR - https://arxiv.org/abs/2606.20497
AB - Navigating the vast space of synthetically accessible molecules demands surrogate models that are interpretable and capable of handling multiple competing objectives at the same time. Deep learning approaches struggle to satisfy them under the computational constraints of quantum-level chemistry. Here, we introduce a modular pipeline that combines interpretable linear meta-learning models and adaptive-confidence uncertainty quantification into an Efficient Global Optimization (EGO) framework for multi-objective molecular discovery. For the first time, linear meta-learning is deployed in a multi-objective chemical search setting: by training across chemical objectives and cheap auxiliary properties, the meta-learned surrogates acquire transferable chemical knowledge that adapts rapidly to new objectives from limited data. Evaluated empirically on a real large scale search for spin-crossover metal-organic complexes, the baseline performs 78% worse in Pareto sense than the meta-learning alternative. We also address the calibration challenges inherent to active search. Since optimal candidates typically lie precisely in the distributional tails, standard uncertainty estimates fail. We introduce an adaptive confidence-tuning algorithm that dynamically recalibrates the exploration-exploitation trade-off as the molecular search evolves. Empirically, dynamic confidence tuning further dominates over 50% of the statically calibrated front.
ER -