E-Book, Englisch, Band 970, 203 Seiten, eBook
Reihe: Lecture Notes in Physics
Helias / Dahmen Statistical Field Theory for Neural Networks
1. Auflage 2020
ISBN: 978-3-030-46444-8
Verlag: Springer International Publishing
Format: PDF
Kopierschutz: 1 - PDF Watermark
E-Book, Englisch, Band 970, 203 Seiten, eBook
Reihe: Lecture Notes in Physics
ISBN: 978-3-030-46444-8
Verlag: Springer International Publishing
Format: PDF
Kopierschutz: 1 - PDF Watermark
Zielgruppe
Graduate
Autoren/Hrsg.
Weitere Infos & Material
Introduction.- Probabilities, moments, cumulants.- Gaussian distribution and Wick’s theorem.- Perturbation expansion.- Linked cluster theorem.- Functional preliminaries.- Functional formulation of stochastic differential equations.- Ornstein-Uhlenbeck process: The free Gaussian theory.- Perturbation theory for stochastic differential equations.- Dynamic mean-field theory for random networks.- Vertex generating function.- Application: TAP approximation.- Expansion of cumulants into tree diagrams of vertex functions.- Loopwise expansion of the effective action - Tree level.- Loopwise expansion in the MSRDJ formalism.- Nomenclature.