Reducing Training Data Needs with Minimal Multilevel Machine Learning (M3L)

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Heinen, Stefan, Khan, Danish, von Rudorff, Guido Falk, Karandashev, Konstantin, Arrieta, Daniel Jose Arismendi, Price, Alastair J. A., Nandi, Surajit, Bhowmik, Arghya, Hermansson, Kersti, von Lilienfeld, O. Anatole
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915275866112000
author Heinen, Stefan
Khan, Danish
von Rudorff, Guido Falk
Karandashev, Konstantin
Arrieta, Daniel Jose Arismendi
Price, Alastair J. A.
Nandi, Surajit
Bhowmik, Arghya
Hermansson, Kersti
von Lilienfeld, O. Anatole
author_facet Heinen, Stefan
Khan, Danish
von Rudorff, Guido Falk
Karandashev, Konstantin
Arrieta, Daniel Jose Arismendi
Price, Alastair J. A.
Nandi, Surajit
Bhowmik, Arghya
Hermansson, Kersti
von Lilienfeld, O. Anatole
contents For many machine learning applications in science, data acquisition, not training, is the bottleneck even when avoiding experiments and relying on computation and simulation. Correspondingly, and in order to reduce cost and carbon footprint, training data efficiency is key. We introduce minimal multilevel machine learning (M3L) which optimizes training data set sizes using a loss function at multiple levels of reference data in order to minimize a combination of prediction error with overall training data acquisition costs (as measured by computational wall-times). Numerical evidence has been obtained for calculated atomization energies and electron affinities of thousands of organic molecules at various levels of theory including HF, MP2, DLPNO-CCSD(T), DFHFCABS, PNOMP2F12, and PNOCCSD(T)F12, and treating tens with basis sets TZ, cc-pVTZ, and AVTZ-F12. Our M3L benchmarks for reaching chemical accuracy in distinct chemical compound sub-spaces indicate substantial computational cost reductions by factors of $\sim$ 1.01, 1.1, 3.8, 13.8 and 25.8 when compared to heuristic sub-optimal multilevel machine learning (M2L) for the data sets QM7b, QM9$^\mathrm{LCCSD(T)}$, EGP, QM9$^\mathrm{CCSD(T)}_\mathrm{AE}$, and QM9$^\mathrm{CCSD(T)}_\mathrm{EA}$, respectively. Furthermore, we use M2L to investigate the performance for 76 density functionals when used within multilevel learning and building on the following levels drawn from the hierarchy of Jacobs Ladder:~LDA, GGA, mGGA, and hybrid functionals. Within M2L and the molecules considered, mGGAs do not provide any noticeable advantage over GGAs. Among the functionals considered and in combination with LDA, the three on average top performing GGA and Hybrid levels for atomization energies on QM9 using M3L correspond respectively to PW91, KT2, B97D, and $τ$-HCTH, B3LYP$\ast$(VWN5), TPSSH.
format Preprint
id arxiv_https___arxiv_org_abs_2308_11196
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Reducing Training Data Needs with Minimal Multilevel Machine Learning (M3L)
Heinen, Stefan
Khan, Danish
von Rudorff, Guido Falk
Karandashev, Konstantin
Arrieta, Daniel Jose Arismendi
Price, Alastair J. A.
Nandi, Surajit
Bhowmik, Arghya
Hermansson, Kersti
von Lilienfeld, O. Anatole
Chemical Physics
For many machine learning applications in science, data acquisition, not training, is the bottleneck even when avoiding experiments and relying on computation and simulation. Correspondingly, and in order to reduce cost and carbon footprint, training data efficiency is key. We introduce minimal multilevel machine learning (M3L) which optimizes training data set sizes using a loss function at multiple levels of reference data in order to minimize a combination of prediction error with overall training data acquisition costs (as measured by computational wall-times). Numerical evidence has been obtained for calculated atomization energies and electron affinities of thousands of organic molecules at various levels of theory including HF, MP2, DLPNO-CCSD(T), DFHFCABS, PNOMP2F12, and PNOCCSD(T)F12, and treating tens with basis sets TZ, cc-pVTZ, and AVTZ-F12. Our M3L benchmarks for reaching chemical accuracy in distinct chemical compound sub-spaces indicate substantial computational cost reductions by factors of $\sim$ 1.01, 1.1, 3.8, 13.8 and 25.8 when compared to heuristic sub-optimal multilevel machine learning (M2L) for the data sets QM7b, QM9$^\mathrm{LCCSD(T)}$, EGP, QM9$^\mathrm{CCSD(T)}_\mathrm{AE}$, and QM9$^\mathrm{CCSD(T)}_\mathrm{EA}$, respectively. Furthermore, we use M2L to investigate the performance for 76 density functionals when used within multilevel learning and building on the following levels drawn from the hierarchy of Jacobs Ladder:~LDA, GGA, mGGA, and hybrid functionals. Within M2L and the molecules considered, mGGAs do not provide any noticeable advantage over GGAs. Among the functionals considered and in combination with LDA, the three on average top performing GGA and Hybrid levels for atomization energies on QM9 using M3L correspond respectively to PW91, KT2, B97D, and $τ$-HCTH, B3LYP$\ast$(VWN5), TPSSH.
title Reducing Training Data Needs with Minimal Multilevel Machine Learning (M3L)
topic Chemical Physics
url https://arxiv.org/abs/2308.11196