BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//modelica.simtek.cc//amfc2026//speaker//8HGL9P
BEGIN:VTIMEZONE
TZID:CST
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1
TZNAME:CST
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-amfc2026-XB79KT@modelica.simtek.cc
DTSTART;TZID=CST:20260922T115500
DTEND;TZID=CST:20260922T122000
DESCRIPTION:This paper introduces State-Invariant Hybrid Models (SIHM)\, a 
 subset of hybrid physics-machine-learning models in which the state vector
  remains defined by the first-principles Modelica model\, while learned co
 mponents contribute missing physics locally. We qualitatively contrast glo
 bal residual learning at system level with local residual learning at comp
 onent level\, arguing that local learning better matches the acausal model
 ing paradigm of Modelica. A tool-agnostic workflow is proposed in which a 
 replaceable dummy component marks the suspected location of missing physic
 s\, while compiler-derived structural information identifies the connected
  states and variables. The workflow is then illustrated in the Wolfram eco
 system on two models to demonstrate the approach's modularity. The results
  indicate that local SIHM components recover targeted missing effects whil
 e preserving the state selection of the original model.
DTSTAMP:20261004T070712Z
LOCATION:Modelica Technology & AI (R1001)
SUMMARY:Learning the Missing Physics Locally with Modelica and Wolfram Lang
 uage - Clément Coïc\, Ankit Naik
URL:https://modelica.simtek.cc/amfc2026/talk/XB79KT/
END:VEVENT
END:VCALENDAR