Cargando…
Differentiable sampling of molecular geometries with uncertainty-based adversarial attacks
Neural network (NN) interatomic potentials provide fast prediction of potential energy surfaces, closely matching the accuracy of the electronic structure methods used to produce the training data. However, NN predictions are only reliable within well-learned training domains, and show volatile beha...
Autores principales: | Schwalbe-Koda, Daniel, Tan, Aik Rui, Gómez-Bombarelli, Rafael |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Nature Publishing Group UK
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8384857/ https://www.ncbi.nlm.nih.gov/pubmed/34429418 http://dx.doi.org/10.1038/s41467-021-25342-8 |
Ejemplares similares
-
Adversarial attacks and adversarial robustness in computational pathology
por: Ghaffari Laleh, Narmin, et al.
Publicado: (2022) -
Improving the adversarial transferability with relational graphs ensemble adversarial attack
por: Pi, Jiatian, et al.
Publicado: (2023) -
Wasserstein Uncertainty Estimation for Adversarial Domain Matching
por: Wang, Rui, et al.
Publicado: (2022) -
Exploiting epistemic uncertainty of the deep learning models to generate adversarial samples
por: Tuna, Omer Faruk, et al.
Publicado: (2022) -
Approaching enzymatic catalysis with zeolites or how to select one reaction mechanism competing with others
por: Ferri, Pau, et al.
Publicado: (2023)