Sökning: onr:"swepub:oai:DiVA.org:kth-345159" >
Representation lear...
Representation learning algorithms for inferring machine independent latent features in pedestals in JET and AUG
-
- Jaervinen, A. E. (författare)
- VTT Tech Res Ctr Finland, FI-02044 Lappeenranta, Finland.;Univ Helsinki, FI-00014 Helsinki, Finland.
-
- Kit, A. (författare)
- VTT Tech Res Ctr Finland, FI-02044 Lappeenranta, Finland.;Univ Helsinki, FI-00014 Helsinki, Finland.
-
- Poels, Y. R. J. (författare)
- Ecole Polytech Fed Lausanne, Swiss Plasma Ctr, CH-1015 Lausanne, Switzerland.;Eindhoven Univ Technol Math & Comp Sci, NL-5600MB Eindhoven, Netherlands.
-
visa fler...
-
- Wiesen, S. (författare)
- Forschungszentrum Julich, Inst Energie & Klimaforschung Plasmaphys, D-52425 Julich, Germany.
-
- Menkovski, V. (författare)
- Eindhoven Univ Technol Math & Comp Sci, NL-5600MB Eindhoven, Netherlands.
-
- Frassinetti, Lorenzo (författare)
- KTH,Elektromagnetism och fusionsfysik
-
- Dunne, M. (författare)
- Max Planck Inst Plasma Phys, D-85748 Garching, Germany.
-
visa färre...
-
VTT Tech Res Ctr Finland, FI-02044 Lappeenranta, Finland;Univ Helsinki, FI-00014 Helsinki, Finland. Ecole Polytech Fed Lausanne, Swiss Plasma Ctr, CH-1015 Lausanne, Switzerland.;Eindhoven Univ Technol Math & Comp Sci, NL-5600MB Eindhoven, Netherlands. (creator_code:org_t)
- AIP Publishing, 2024
- 2024
- Engelska.
-
Ingår i: Physics of Plasmas. - : AIP Publishing. - 1070-664X .- 1089-7674. ; 31:3
- Relaterad länk:
-
https://doi.org/10.1...
-
visa fler...
-
https://urn.kb.se/re...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- Variational autoencoder (VAE)-based representation learning algorithms are explored for their capability to disentangle tokamak size dependence from other dependencies in a dataset of thousands of observed pedestal electron density and temperature profiles from JET and ASDEX Upgrade tokamaks. Representation learning aims to establish a useful representation that characterizes the dataset. In the context of magnetic confinement fusion devices, a useful representation could be considered to map the high-dimensional observations to a manifold that represents the actual degrees of freedom of the plasma scenario. A desired property for these representations is organization of the information into disentangled variables, enabling interpretation of the latent variables as representations of semantically meaningful characteristics of the data. The representation learning algorithms in this work are based on VAE that encodes the pedestal profile information into a reduced dimensionality latent space and learns to reconstruct the full profile information given the latent representation. Attaching an auxiliary regression objective for the machine control parameter configuration, broadly following the architecture of the domain invariant variational autoencoder (DIVA), the model learns to associate device control parameters with the latent representation. With this multimachine dataset, the representation does encode density scaling with device size that is qualitatively consistent with Greenwald density limit scaling. However, if the major radius of the device is given through a common regression objective with the other machine control parameters, the latent state of the representation struggles to clearly disentangle the device size from changes of the other machine control parameters. When separating the device size as an independent latent variable with dedicated regression objectives, similar to separation of domain and class labels in the original DIVA publication, the latent space becomes well organized as a function of the device size.
Ämnesord
- NATURVETENSKAP -- Fysik (hsv//swe)
- NATURAL SCIENCES -- Physical Sciences (hsv//eng)
Publikations- och innehållstyp
- ref (ämneskategori)
- art (ämneskategori)
Hitta via bibliotek
Till lärosätets databas