BEGIN:VCALENDAR
VERSION:2.0
METHOD:PUBLISH
CALSCALE:GREGORIAN
PRODID:-//WordPress - MECv6.5.6//EN
X-ORIGINAL-URL:https://cmc.deusto.eus/
X-WR-CALNAME:cmc.deusto.eus
X-WR-CALDESC:DeustoCCM - Chair of Computational Mathematics at University of Deusto
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-PUBLISHED-TTL:PT1H
X-MS-OLK-FORCEINSPECTOROPEN:TRUE
BEGIN:VEVENT
CLASS:PUBLIC
UID:MEC-482018fcb72d6725497bebbe9bb83833@cmc.deusto.eus
DTSTART:20230420T090000Z
DTEND:20230420T100000Z
DTSTAMP:20251031T213100Z
CREATED:20251031
LAST-MODIFIED:20251031
PRIORITY:5
TRANSP:OPAQUE
SUMMARY:Error bounds for physics-informed (and) operator learning for PDEs
DESCRIPTION:Next Thursday April 20, 2023:\nOrganized by: FAU DCN-AvH, Chair for Dynamics, Control, Machine Learning and Numerics – Alexander von Humboldt Professorship at FAU, Friedrich-Alexander-Universität Erlangen-Nürnberg (Germany)\nTitle: Error bounds for physics-informed (and) operator learning for PDEs\nSpeaker: Tim De Ryck\nAffiliation: PhD Student from ETH Zürich\nAbstract. In recent years, many deep learning architectures have been proposed to approximate solutions to partial differential equations (PDEs). Despite their impressive performance in experiments, a corresponding mathematical theory is lacking. \nWe propose a very general framework for deriving rigorous bounds on the approximation error for physics-informed neural networks (PINNs) and operator learning architectures such as DeepONets and FNOs as well as for physics-informed operator learning. These bounds guarantee that PINNs and (physics-informed) DeepONets or FNOs will efficiently approximate the underlying solution or solution operator of generic PDEs. [1]\nHowever, PINNs and related architectures require regularity of the underlying PDE solution to guarantee accurate approximation and may fail at approximating discontinuous solutions of PDEs such as nonlinear hyperbolic equations. To ameliorate this, we propose a novel variant of PINNs, termed as weak PINNs (wPINNs) for accurate approximation of entropy solutions of scalar conservation laws. We prove rigorous bounds on the error incurred by wPINNs and illustrate their performance through numerical experiments to demonstrate that wPINNs can approximate entropy solutions accurately. [2]\nJoint work with Prof. Dr. Siddhartha Mishra (ETH Zürich) and Roberto Molinaro (ETH Zürich). \nWHERE?\nOn-site / Online\nOn-site:\nRoom Übung 3 | 01.252-128\n1st. floor. Department Mathematik. Friedrich-Alexander-Universität Erlangen-Nürnberg\nCauerstraße 11, 91058 Erlangen\nGPS-Koord. Raum: 49.573572N, 11.030394E\nOnline:\nZoom meeting link\nMeeting ID: 614 4658 159 | PIN code: 914397\nThis event on LinkedIn\n[1] De Ryck, Tim, and Siddhartha Mishra. “Generic bounds on the approximation error for physics-informed (and) operator learning.” NeurIPS (2022). \n[2] De Ryck, Tim, Siddhartha Mishra, and Roberto Molinaro. “wPINNs: Weak physics-informed neural networks for approximating entropy solutions of hyperbolic conservation laws.” Preprint (2022). \n
URL:https://cmc.deusto.eus/events-calendar/error-bounds-for-physics-informed-and-operator-learning-for-pdes/
ORGANIZER;CN=FAU DCN-AvH:MAILTO:
CATEGORIES:FAU DCN-AvH Jr. Seminar,Seminar/Talk
LOCATION:FAU - Faculty of Sciences
ATTACH;FMTTYPE=image/png:https://cmc.deusto.eus/wp-content/uploads/FAUDCNAvHJrSeminar_tDeRyck_20apr2023.png
END:VEVENT
END:VCALENDAR