Using principal component analysis for neural network high-dimensional potential energy surface

Bastien Casier, S. Carniato, Tsveta Miteva, Nathalie Capron, Nicolas Sisourat

The Journal of Chemical Physics 2020 article cited by 14

Source Open-access PDF More in Physics & Astronomy

importance 5/10 ai assisted Machine Learning in Materials Science computational chemistry stage: research

Problem

Computing high-dimensional potential energy surfaces for chemical reactions remains difficult for most relevant systems with electronic structure methods.

Why it matters

Potential energy surfaces underpin understanding and prediction of chemical reactions.

What the AI did

PCA prepares an optimal, reduced set of molecular descriptors for a neural network that learns and predicts the potential energy surface.

Results

Substantial improvement in NN learning/prediction and descriptor reduction without accuracy loss; no numeric metrics stated.

Impact

Demonstrated on keto–enol tautomerization PES; improvement described only qualitatively.

Record

Venue	The Journal of Chemical Physics
Published	2020
Work type	article
Authors	Bastien Casier, S. Carniato, Tsveta Miteva, Nathalie Capron, Nicolas Sisourat
Citations	14 (OpenAlex, at import time)
References	43
Model or method	artificial neural network with PCA-prepared descriptors
Organization	not stated in source
Primary source	https://doi.org/10.1063/5.0009264
Open access	bronze
Keywords	potential energy surface, neural network, PCA, molecular descriptors, keto-enol tautomerization

Importance rating

5 / 10

real but narrower advance

PESs are described as central to understanding chemical reactions, a foundational chemistry task, but demonstration is limited to one molecule (acetone keto-enol tautomerism); no scale given.

Basis: inferred. Judged from the source text, which gives no explicit scale. Stated scale: not stated in source. Method and limits.

More in Physics & Astronomy

#	Paper	Year	Citations	Importance
1	Magnetic control of tokamak plasmas through deep reinforcement learning Jonas Degrave, Federico Felici, Jonas Buchli, Michael Neunert, Brendan...	2022	739	7
2	Solving the quantum many-body problem with artificial neural networks Giuseppe Carleo, Matthias Troyer	2017	2,205	7
3	Learning high-accuracy error decoding for quantum processors Johannes Bausch, Andrew Senior, Francisco J. H. Heras, Thomas Edlich,...	2024	90	7
4	Optimizing Quantum Error Correction Codes with Reinforcement Learning Hendrik Poulsen Nautrup, Nicolas Delfosse, Vedran Dunjko, Hans J. Brie...	2019	158	7
5	Machine-learning-assisted correction of correlated qubit errors in a topological code Paul Baireuther, Thomas E. O’Brien, Brian Tarasinski, C. W. J. Beenakk...	2018	141	7