Research

Research Interests

My research focuses on developing robust and efficient machine learning methods with real-world applicability. Specific areas of interest include:

Representation Learning – learning compact, transferable feature representations from limited labelled data.
Uncertainty Quantification – building models that know what they do not know.
Scientific Machine Learning – applying ML to accelerate discovery in physics, biology, and materials science.
Explainability & Fairness – ensuring models are interpretable and unbiased.

[Title of Paper 1]
Nakul Padalkar, Co-Author A, Co-Author B
Conference/Journal Name, 2024
[PDF] [Code] [Slides]

Brief abstract or one-sentence description of the contribution goes here.

[Title of Paper 2]
Nakul Padalkar, Co-Author C
Workshop on Topic X at Conference Y, 2024
[PDF] [Poster]

Brief abstract or one-sentence description of the contribution goes here.

[Title of Paper 3]
Co-Author D, Nakul Padalkar, Co-Author E
Journal Name, 2023
[PDF] [Code]

Brief abstract or one-sentence description of the contribution goes here.

[Title of Preprint]
Nakul Padalkar, Co-Author F
arXiv, 2024
[arXiv]

Brief description of the preprint.

Project	Description	Links
Project Alpha	Efficient neural architecture search for scientific domains.	GitHub
Project Beta	Benchmark dataset and evaluation suite for uncertainty estimation.	GitHub

Project	Description	Links
Project Gamma	Reproducible ML pipelines using Quarto and DVC.	GitHub

I am fortunate to collaborate with researchers across several institutions. If you are interested in collaborating, please feel free to get in touch.