English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Matbench Discovery simulates the deployment of machine learning (ML) energymodels in a high-throughput search for stable inorganic crystals. We addressthe disconnect between (i) thermodynamic stability and formation energy and(ii) in-domain vs out-of-distribution performance. Alongside this paper, wepublish a Python package to aid with future model submissions and a growingonline leaderboard with further insights into trade-offs between variousperformance metrics. To answer the question which ML methodology performs bestat materials discovery, our initial release explores a variety of modelsincluding random forests, graph neural networks (GNN), one-shot predictors,iterative Bayesian optimizers and universal interatomic potentials (UIP).Ranked best-to-worst by their test set F1 score on thermodynamic stabilityprediction, we find CHGNet > M3GNet > MACE > ALIGNN > MEGNet > CGCNN > CGCNN+P> Wrenformer > BOWSR > Voronoi tessellation fingerprints with random forest.The top 3 models are UIPs, the winning methodology for ML-guided materialsdiscovery, achieving F1 scores of ~0.6 for crystal stability classification anddiscovery acceleration factors (DAF) of up to 5x on the first 10k most stablepredictions compared to dummy selection from our test set. We also highlight asharp disconnect between commonly used global regression metrics and moretask-relevant classification metrics. Accurate regressors are susceptible tounexpectedly high false-positive rates if those accurate predictions lie closeto the decision boundary at 0 eV/atom above the convex hull where mostmaterials are. Our results highlight the need to focus on classificationmetrics that actually correlate with improved stability hit rate.

Matbench Discovery -- An evaluation framework for machine learning crystal stability prediction