Adaptive Uncertainty-Aware Deep Learning for Materials Discovery With High-Dimensional Design Inputs
Bibliographic record
Abstract
Abstract High-dimensional structure and composition spaces present a major challenge in materials discovery due to the difficulty of efficiently navigating vast and complex design space. Additionally, most existing machine learning approaches lack the capability to quantify epistemic uncertainty, which arises from limited data, a critical limitation for materials discovery tasks involving high-dimensional representations such as atomic structures. To address these challenges, we introduce UPNet, an uncertainty-aware atomistic machine learning model within a Bayesian Optimization (BO) framework. UPNet enables automated representation learning directly from high-dimensional atomic structures while providing principled uncertainty quantification through the use of Spectral-normalized Neural Gaussian Process (SNGP). By incorporating a constrained expected improvement acquisition function, our BO framework optimizes multiple evaluation criteria simultaneously. We demonstrate the effectiveness of our approach in catalyst discovery for the CO2 reduction reaction, where it achieves high prediction accuracy, facilitates interpretable feature extraction, and enables multicriteria design optimization. Our method reduces computational cost and time, achieving a 10× reduction in the number of required simulation calculations. Beyond catalysis, this framework offers a broadly applicable solution for accelerating materials discovery in various domains that involve high-dimensional design inputs and expensive physics-based simulations.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".