Manifolds of quasi-constant SOAP and ACSF fingerprints and the resulting failure to machine learn four-body interactions.

@article{Parsaeifard2022ManifoldsOQ,
  title={Manifolds of quasi-constant SOAP and ACSF fingerprints and the resulting failure to machine learn four-body interactions.},
  author={Behnam Parsaeifard and Stefan Goedecker},
  journal={The Journal of chemical physics},
  year={2022},
  volume={156 3},
  pages={
          034302
        }
}
Atomic fingerprints are commonly used for the characterization of local environments of atoms in machine learning and other contexts. In this work, we study the behavior of two widely used fingerprints, namely, the smooth overlap of atomic positions (SOAP) and the atom-centered symmetry functions (ACSFs), under finite changes of atomic positions and demonstrate the existence of manifolds of quasi-constant fingerprints. These manifolds are found numerically by following eigenvectors of the… 

Figures from this paper

GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations
We present our latest advancements of machine-learned potentials (MLPs) based on the neuroevolution potential (NEP) framework introduced in [Fan et al. , Phys. Rev. B 104 , 104309 (2021)] and their
Graph Neural Network Potential for Magnetic Materials
Machine Learning (ML) interatomic potential has shown its great power in condensed matter physics. However, ML interatomic potential for a magnetic system including both structural degrees of freedom
Local invertibility and sensitivity of atomic structure-feature mappings
Background: The increasingly common applications of machine-learning schemes to atomic-scale simulations have triggered efforts to better understand the mathematical properties of the mapping between
Pointwise distance distributions of periodic point sets
TLDR
The new invariants that are crystal descriptors without false negatives and are called Pointwise Distance Distributions (PDD) are introduced, which are numerical matrices with a near-linear time complexity and an exactly computable metric.

References

SHOWING 1-10 OF 31 REFERENCES
‘G’
  • P. Alam
  • Composites Engineering: An A–Z Guide
  • 2021
[C]
  • Thomas de Quincey
  • Physics
    The Works of Thomas De Quincey, Vol. 1: Writings, 1799–1820
  • 2000
In supernova (SN) spectroscopy relatively little attention has been given to the properties of optically thick spectral lines in epochs following the photosphere’s recession. Most treatments and
‘A’
  • P. Alam
  • Medicine
    Composites Engineering: An A–Z Guide
  • 2021
TLDR
A fluorescence-imaging-based endoscopic capsule that automates the detection process of colorectal cancer was designed and developed in the lab and offered great possibilities for future applicability in selective and specific detection of other fluorescently labelled cancers.
Local invertibility and sensitivity of atomic structure-feature mappings
Background: The increasingly common applications of machine-learning schemes to atomic-scale simulations have triggered efforts to better understand the mathematical properties of the mapping between
and a at
The xishacorene natural products are structurally unique apolar diterpenoids that feature a bicyclo[3.3.1] framework. These secondary metabolites likely arise from the well-studied, structurally
In Advances in Neural Information Processing Systems 15
The "DNA" of chemistry: Scalable quantum machine learning with "amons"
TLDR
The use of amons affords highly accurate machine learning predictions of quantum properties of arbitrary query molecules in real time for predicting energies of various covalently and non-covalently bonded systems.
“A and B”:
Direct fabrication of large micropatterned single crystals. p1205 21 Feb 2003. (news): Academy plucks best biophysicists from a sea of mediocrity. p994 14 Feb 2003.
Phys
  • Rev. B 98, 144102
  • 2018
Phys
  • Rev. Lett. 109, 059801
  • 2012
...
...