• New Preprint: FEKAN: Feature-Enriched Kolmogorov-Arnold Networks [Arxiv]. 

We present an efficient and accurate variant of KAN, and compare its performance with state-of-the-art KAN models and their fast variants. Congratulations to Ph.D. student Sidharth S. Menon.

• New Preprint: In Transformer We Trust? A Perspective on Transformer Architecture Failure Modes [Arxiv]. Congratulations to Ph.D. student Trishit Mondal.

• New Preprint: Hypersonic Flow Control: Generalized Deep Reinforcement Learning for Hypersonic Intake Unstart Control under Uncertainty. [Arxiv]. Congratulations to Ph.D. student Trishit Mondal on this excellent work.

• Congratulations to Ph.D students Sidharth S. Menon (for his third publication) and Trishit Mondal (for his first publication) in Neural Networks, Elsevier.

In this work, we propose a rigorous definitions and a comprehensive applications for Scientific Foundation Models:

S. S. Menon, T. Mondal, S. Brahmachary, A. Panda, S. M. Joshi, K. Kalyanaraman, Ameya D. Jagtap, On Scientific Foundation Models: Rigorous Definitions, Key Applications, and a Comprehensive Survey, Neural Networks, Elsevier, 198 (2026) 108567. [Journal]

• New Preprint: A comprehensive survey of Deep Learning methods for fluid flows (Turbulent, Multiphase, and Reactive Flows)

S.S. Menon, M. Lavari, A. Kokernak, J. Mathew, C. Jagtap, J. Jayachandran, A. Gnanaskandan, and Ameya D. Jagtap, Intelligent Fluid Flows: A Survey of Deep Learning Methods for Turbulent, Multiphase, and Reactive Flows. Available at SSRN: https://ssrn.com/abstract=5907058 or http://dx.doi.org/10.2139/ssrn.5907058 [Link]

• Congratulations to Ph.D students Yunhao Zhang and Sidharth S. Menon on their first co-author publication in Computer Methods in Applied Mechanics and Engineering, Elsevier.

Y. Zhang, S. S. Menon L. Cheng, A. Gnanaskandan, Ameya D. Jagtap, BubbleOKAN: A Physics-Informed Interpretable Neural Operator for High-Frequency Bubble Dynamics, Computer Methods in Applied Mechanics and Engineering, Volume 450, (2026) 118667. [Journal]

This paper presents BubbleONet, a physics-informed operator learning surrogate for efficiently simulating high-frequency bubble dynamics governed by Rayleigh–Plesset and Keller–Miksis equations.

• Upcoming talk at ACSEL Technologies Forum, India, Jan. 23,  2026: Scientific Machine Learning: Where We Stand and Where the Future Leads by Dr. A. D. Jagtap. (Plenary Talk) 

• Congratulations to Ph.D. student Trishit Mondal on his first preprint:

Trishit Mondal, Ricardo Vinuesa, and Ameya D. Jagtap, Shocks Under Control: Taming Transonic Compressible Flow over an RAE2822 Airfoil with Deep Reinforcement Learning [Arxiv]

Great work Trishit!

• Recognized Among the World’s Top 2% Scientists by Stanford and Elsevier in 2024 and 2025: In 2025, Dr. Ameya D. Jagtap was once again recognized among the world’s top 2% scientists by Stanford University and Elsevier, with his global rank improving from 203 last year to 112 this year.

• Congratulations to Ph.D. student Sidharth S. Menon on his first publication in Computer Methods in Applied Mechanics and Engineering (Elsevier)

Sidharth S. Menon and Ameya D. Jagtap, Anant-Net: Breaking the curse of dimensionality with scalable and interpretable neural surrogate for high-dimensional PDEs, Computer Methods in Applied Mechanics and Engineering, Volume 447, (2025) 118403. [Journal] [Arxiv]25

• Paper accepted in Neurocomputing, Elsevier: Jassem Abbasi, Ameya D. Jagtap, Ben Moseley, Aksel Hiorth, and Pal Østebø Andersen, Challenges and Advancements in Modeling Shock Fronts with Physics-Informed Neural Networks: A Review and Benchmarking Study, Neurocomputing, 657 (2025) 131440. [Journal] [Arxiv]

• New Preprint: An Approximation Theory Perspective on Machine Learning. Our goal is to spark a more rigorous and inspiring theoretical dialogue, one that places approximation theory at the heart of modern ML. Here is the link [Arxiv].