AG

January 1, 2021by Dr Abhinav Gupta

Scientific Machine Learning

What attracted me to scientific machine learning was not the idea of replacing physics, but the possibility of accelerating expensive simulations while still respecting physical laws. I saw it as another step toward the larger goal that had been driving me for years: reducing computational cost without compromising predictive capability.

Scientific Machine Learning (SciML)

Neural operators, surrogate models, and physics-informed learning introduced new pathways for accelerating simulations and enabling real-time or near real-time engineering predictions. This perspective continues to shape my current research in computational mechanics and AI-driven scientific computing.