Department of Computational and Data Sciences
Department Seminar
Speaker : Dr. Prashant K. Jha, Assistant Professor, USA
Title : Error Correction and Adaptation for Reliable Neural Operators
Date & Time : October 22nd, 2026 (Thursday), 04:00 PM
Venue : # 102, CDS Seminar Hall
ABSTRACT:
Neural operators offer fast approximations of PDE solution maps, but their accuracy can deteriorate as optimization, inference, or changing operating conditions take them beyond their training data. Their use therefore requires assessing individual predictions and 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 corrector improves a neural-operator prediction through a linearized variational solve, without retraining. Under suitable regularity and stability assumptions, it yields local quadratic error reduction. Examples in PDE-constrained optimization and inference will illustrate its benefits, followed by the challenges of learning the correction itself.
I will then present computational-twin studies of magnetically actuated soft structures under changing stiffness and loading. These studies combine prediction-specific error assessment with selective finite element evaluations and reuse the acquired solution paths to adapt the operator and its estimator. Adaptation improves predictions and subsequent magnetic-command selection, but improved forward accuracy does not consistently produce reliable error assessment. The talk will conclude with a discussion of what these results establish, where the methods remain limited, and the questions they raise for reliable scientific machine learning.
BIOGRAPHY:
Prashant 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 Carnegie Mellon University and subsequently held postdoctoral positions at Louisiana State University and the Oden Institute at The University of Texas at Austin. His work in scientific machine learning focuses on neural operators, 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 predictions for simulation, optimization, and inference
Host Faculty: Prof. Phani Motamarri
ALL ARE WELCOME
