Seminar

In today’s era of large-scale simulations and experiments, the demand for efficient computational models has seen significant growth.

Surrogate modeling, a technique used to approximate expensive models or simulations with simpler, data-driven models, plays an essential role in this arena.

This presentation focuses on the most recent advancements in data-driven surrogate modeling, highlighting novel methodologies and their implications for various scientific and engineering domains.
The presentation begins with providing a brief overview of surrogate modeling concept, setting the stage for newer data-driven approaches.

Recent trends point towards the incorporation of machine learning and artificial intelligence in crafting surrogate models, tapping into their ability to glean complex patterns from vast datasets.

Emphasis will be placed on cutting-edge techniques such as deep learning-based surrogates, Gaussian process regression, and advancements in multi-fidelity surrogate modeling.

To wrap up, the presentation will discuss the challenges and opportunities ahead for data-driven surrogate modeling, emphasizing the ongoing research areas and the potential future directions for this promising field.