Vignesh Sella

I am an AI/ML Resident at Google X, the Moonshot Factory, and a PhD candidate in the Computational Science, Engineering, and Mathematics program at The University of Texas at Austin, where I am advised by Karen Willcox and Anirban Chaudhuri at the Oden Institute for Computational Engineering and Sciences.

My research develops data-efficient machine learning methods for science and engineering problems where training data is expensive and scarce. The central challenge is one that extends well beyond any single application domain: when learning from data is costly, how do you make the most of what you have? My work addresses this through multifidelity surrogate modeling, which leverages correlations between cheap approximate models and limited high-fidelity data to construct more efficient empirical risk estimators. This approach has broad relevance to any setting where reducing the cost of learning from data matters, thus, I am always excited to explore new applications and would be happy to chat about potential collaborations.

Outside of research, I am a lifelong football (soccer) player and fan. Força Barça! I also love to watch movies and prestige TV, and I try to travel as much as possible, having been fortunate enough to explore many corners of the world.

news

May 04, 2026	Officially a PhD Candidate!
Dec 11, 2025	Published a journal article in the Springer Nature Journal of Machine Learning for Computational Science and Engineering special issue.
Jan 27, 2025	Joining X, the Moonshot Factory as an AI/ML Resident!

latest posts

Jan 27, 2025	Joining Google X, the Moonshot Factory as an AI Resident

selected publications

FDS
Multifidelity linear regression for scientific machine learning from scarce data

Elizabeth Qian, Dayoung Kang, Vignesh Sella, and 1 more author

Foundations of Data Science, 2024

Abs DOI arXiv Bib HTML PDF

We propose a new multifidelity training approach for scientific machine learning that exploits scientific contexts where data of varying fidelities and costs are available. We use multifidelity data to define new multifidelity Monte Carlo estimators for the unknown parameters of linear regression models, providing theoretical analyses that guarantee accuracy and improved robustness to small training budgets. Numerical results demonstrate that multifidelity learned models achieve order-of-magnitude lower model variance than standard models trained on only high-fidelity data of comparable cost.
@article{qian2024multifidelity, title = {Multifidelity linear regression for scientific machine learning from scarce data}, author = {Qian, Elizabeth and Kang, Dayoung and Sella, Vignesh and Chaudhuri, Anirban}, journal = {Foundations of Data Science}, publisher = {American Institute of Mathematical Sciences}, year = {2024}, doi = {10.3934/fods.2024049}, }
AIAA
Improving Neural Network Efficiency With Multifidelity and Dimensionality Reduction Techniques

Vignesh Sella, Thomas O’Leary-Roseberry, Xiaosong Du, and 5 more authors

In AIAA SciTech Forum, 2025

Abs DOI Bib HTML PDF

This work utilizes output dimensionality reduction along with information from multiple models of varying fidelities and cost to develop accurate projection-enabled multifidelity neural networks (MF-NNs) with a limited computational budget. Three approaches for MF-NNs leveraging proper orthogonal decomposition based projections are introduced: pre-training, additive, and multi-step. The MF-NN is applied to approximate the optimal design of 2D aerodynamic airfoils given performance and design requirements, demonstrating significant improvements over single-fidelity neural networks in data-scarce regimes.
@inproceedings{sella2025improving, title = {Improving Neural Network Efficiency With Multifidelity and Dimensionality Reduction Techniques}, author = {Sella, Vignesh and O'Leary-Roseberry, Thomas and Du, Xiaosong and Guo, Mengwu and Martins, Joaquim and Ghattas, Omar and Willcox, Karen and Chaudhuri, Anirban}, booktitle = {AIAA SciTech Forum}, year = {2025}, address = {Orlando, FL}, doi = {10.2514/6.2025-2807}, }
MLCSE
Projection-based multifidelity linear regression for data-scarce applications

Vignesh Sella, Julie Pham, Karen Willcox, and 1 more author

Machine Learning for Computational Science and Engineering, 2025

Abs DOI arXiv Bib

This work develops projection-based multifidelity linear regression methods for data-scarce applications with high-dimensional outputs. By combining high-fidelity and low-fidelity model evaluations with proper orthogonal decomposition projections, the approach achieves improved predictive accuracy over single-fidelity methods in low-data regimes. Theoretical and numerical results demonstrate robustness and accuracy improvements for surrogate modeling in computational engineering applications.
@article{sella2025projection, title = {Projection-based multifidelity linear regression for data-scarce applications}, author = {Sella, Vignesh and Pham, Julie and Willcox, Karen and Chaudhuri, Anirban}, journal = {Machine Learning for Computational Science and Engineering}, volume = {1}, number = {2}, pages = {47}, year = {2025}, publisher = {Springer}, doi = {10.1007/s44379-025-00049-5}, }