BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.4.8//EN
TZID:Asia/Kolkata
X-WR-TIMEZONE:Asia/Kolkata
BEGIN:VEVENT
UID:230@cds.iisc.ac.in
DTSTART;TZID=Asia/Kolkata:20261022T160000
DTEND;TZID=Asia/Kolkata:20261022T170000
DTSTAMP:20261019T112542Z
URL:https://cds.iisc.ac.in/events/seminar-cds-102-october-22nd-1600-error-
 correction-and-adaptation-for-reliable-neural-operators/
SUMMARY:{Seminar} @ CDS: #102\, October 22nd: 16:00: "Error Correction and 
 Adaptation for Reliable Neural Operators."
DESCRIPTION:Department of Computational and Data Sciences\nDepartment Semin
 ar\n\n\n\nSpeaker : Dr. Prashant K. Jha\, Assistant Professor\, USA\nTitle
  : Error Correction and Adaptation for Reliable Neural Operators\nDate &am
 p\; Time : October 22nd\, 2026 (Thursday)\, 04:00 PM\nVenue : # 102\, CDS 
 Seminar Hall\n\n\n\nABSTRACT:\nNeural operators offer fast approximations 
 of PDE solution maps\, but their accuracy can deteriorate as optimization\
 , inference\, or changing operating conditions take them beyond their trai
 ning data. Their use therefore requires assessing individual predictions a
 nd understanding how approximation errors affect the quantities used in a 
 design or decision. I will discuss our work on these questions\, beginning
  with goal-oriented error estimation and residual-based correction. The co
 rrector improves a neural-operator prediction through a linearized variati
 onal solve\, without retraining. Under suitable regularity and stability a
 ssumptions\, it yields local quadratic error reduction. Examples in PDE-co
 nstrained optimization and inference will illustrate its benefits\, follow
 ed by the challenges of learning the correction itself.\n\nI will then pre
 sent computational-twin studies of magnetically actuated soft structures u
 nder changing stiffness and loading. These studies combine prediction-spec
 ific error assessment with selective finite element evaluations and reuse 
 the acquired solution paths to adapt the operator and its estimator. Adapt
 ation improves predictions and subsequent magnetic-command selection\, but
  improved forward accuracy does not consistently produce reliable error as
 sessment. The talk will conclude with a discussion of what these results e
 stablish\, where the methods remain limited\, and the questions they raise
  for reliable scientific machine learning.\n\nBIOGRAPHY:\nPrashant K. Jha 
 is an Assistant Professor of Mechanical Engineering at South Dakota Mines\
 , where he leads the Computational Engineering Analysis and Design Lab. He
  earned his M.E. at the Indian Institute of Science and his Ph.D. at Carne
 gie Mellon University and subsequently held postdoctoral positions at Loui
 siana State University and the Oden Institute at The University of Texas a
 t Austin. His work in scientific machine learning focuses on neural operat
 ors\, error estimation and correction\, and adaptive computational methods
 \, with applications in computational mechanics. He combines mathematical 
 analysis and numerical methods to improve the reliability of data-driven p
 redictions for simulation\, optimization\, and inference\n\nHost Faculty: 
 Prof. Phani Motamarri\n\n\n\nALL ARE WELCOME
CATEGORIES:Events,Talks
END:VEVENT
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
X-LIC-LOCATION:Asia/Kolkata
BEGIN:STANDARD
DTSTART:20251022T160000
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
END:STANDARD
END:VTIMEZONE
END:VCALENDAR