Sökning: onr:"swepub:oai:DiVA.org:uu-498148" >
Fortnet, a software...
Fortnet, a software package for training Behler-Parrinello neural networks
-
- van der Heide, T. (författare)
- Univ Bremen, Bremen Ctr Computat Mat Sci, Bremen, Germany.
-
- Kullgren, Jolla, 1978- (författare)
- Uppsala universitet,Strukturkemi
-
- Broqvist, Peter (författare)
- Uppsala universitet,Strukturkemi
-
visa fler...
-
- Bacic, V. (författare)
- Jacobs Univ Bremen, Dept Phys & Earth Sci, Campus Ring 1, D-28759 Bremen, Germany.
-
- Frauenheim, T. (författare)
- Univ Bremen, Bremen Ctr Computat Mat Sci, Bremen, Germany.;Beijing Computat Sci Res Ctr, Beijing 100193, Peoples R China.;Shenzhen JL Computat Sci & Appl Res Inst CSAR, Shenzhen 518110, Peoples R China.
-
- Aradi, B. (författare)
- Univ Bremen, Bremen Ctr Computat Mat Sci, Bremen, Germany.
-
visa färre...
-
Univ Bremen, Bremen Ctr Computat Mat Sci, Bremen, Germany Strukturkemi (creator_code:org_t)
- Elsevier, 2023
- 2023
- Engelska.
-
Ingår i: Computer Physics Communications. - : Elsevier. - 0010-4655 .- 1879-2944. ; 284
- Relaterad länk:
-
https://doi.org/10.1...
-
visa fler...
-
https://uu.diva-port... (primary) (Raw object)
-
https://urn.kb.se/re...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- A new, open source, parallel, stand-alone software package (Fortnet) has been developed, which implements Behler-Parrinello neural networks. It covers the entire workflow from feature generation to the evaluation of generated potentials, coupled with higher-level analysis such as the analytic calculation of atomic forces. The functionality of the software package is demonstrated by driving the training for the fitted correction functions of the density functional tight binding (DFTB) method, which are commonly used to compensate the inaccuracies resulting from the DFTB approximations to the Kohn -Sham Hamiltonian. The usual two-body form of those correction functions limits the transferability of the parametrizations between very different structural environments. The recently introduced DFTB+ANN approach strives to lift these limitations by combining DFTB with a near-sighted artificial neural network (ANN). After investigating various approaches, we have found the combination of DFTB with an ANN acting on-top of some baseline correction functions (delta learning) the most accurate one. It allowed to introduce many-body corrections on top of two-body parametrizations, while excellent transferability to chemical environments with deviating energetics could be demonstrated.
Ämnesord
- NATURVETENSKAP -- Kemi -- Fysikalisk kemi (hsv//swe)
- NATURAL SCIENCES -- Chemical Sciences -- Physical Chemistry (hsv//eng)
- NATURVETENSKAP -- Fysik -- Den kondenserade materiens fysik (hsv//swe)
- NATURAL SCIENCES -- Physical Sciences -- Condensed Matter Physics (hsv//eng)
Nyckelord
- Fortnet
- Fortran
- Neural networks
- BPNN
- DFTB
- Many -body repulsive
Publikations- och innehållstyp
- ref (ämneskategori)
- art (ämneskategori)
Hitta via bibliotek
Till lärosätets databas