Sanketh Vedula

I’m a PhD student at the Computer Science Department, Technion – Israel Institute of Technology advised by Alex Bronstein.

I am interested in developing machine learning methods that accelerate scientific discovery. In particular, I research and develop tools involving generative modeling, optimal transport, geometric deep learning, and distribution-free uncertainty estimation, to tackle problems in computational imaging, computational and structural biology, computational chemistry, and physics.

I aspire to build solutions for challenging problems that have real-world impact. I love science. I enjoy doing research, learning, and discussing ideas with people.

In parallel to my graduate studies, I was an ML Researcher at Sibylla from 2020-23, where I built mid-frequency quant-trading and portfolio management strategies. In 2024, I spent 8 months as an ML Research Scientist intern at Pfizer, Berlin working on problems in pharmaceutical chemistry and structural biology with Djork-Arné Clevert, Kristof Schütt, and Joren Retel.

I received an M.Sc. (cum laude) in Computer Science from Technion in 2020, where I was advised by Alex Bronstein and Michael Zibulevsky. Earlier, I received a B.Eng. (Hons.) in Computer Science from BITS Pilani in 2017. I also held visiting positions at Institute of Science & Technology Austria, Imperial College London, and Chennai Mathematical Institute.

news

Jun 1, 2024	Preprint on scalable Gromov-Wasserstein solvers
May 1, 2024	Technion-Pfizer Fellowship
Jan 14, 2024	Manifold VQR at AISTATS 2024
Jan 1, 2024	Visiting IST Austria
Oct 28, 2023	Awarded excellence scholarship by ISTRC (2023-24)
Oct 20, 2023	Two papers accepted at VLDB 2024
Jun 19, 2023	Two papers accepted to ICML Workshops
Feb 7, 2023	GRACE at ASPLOS 2023

selected publications

ICML Workshop

Scalable unsupervised alignment of general metric and non-metric structures

S. Vedula, V. Maiorca, L. Basile, and 2 more authors

arXiv preprint arXiv:2406.13507 (AI for Science Workshop, ICML), 2024

Bib

@article{vedula2024scalable,
  title = {Scalable unsupervised alignment of general metric and non-metric structures},
  author = {Vedula, S. and Maiorca, V. and Basile, L. and Locatello, F. and Bronstein, A.},
  journal = {arXiv preprint arXiv:2406.13507 (AI for Science Workshop, ICML)},
  year = {2024},
}

bioRxiv

Seeing Double: Molecular dynamics simulations reveal the stability of certain alternate protein conformations in crystal structures

A. Rosenberg, S. Vedula, A. Bronstein, and 1 more author

bioRxiv, 2024

Abs

Proteins jiggle around, adopting ensembles of interchanging conformations. Here we show through a large-scale analysis of the Protein Data Bank and using molecular dynamics simulations, that segments of protein chains can also commonly adopt dual, transiently stable conformations which is not explained by direct interactions. Our analysis highlights how alternate conformations can be maintained as non-interchanging, separated states intrinsic to the protein chain, namely through steric barriers or the adoption of transient secondary structure elements. We further demonstrate that despite the commonality of the phenomenon, current structural ensemble prediction methods fail to capture these bimodal distributions of conformations.Competing Interest StatementThe authors have declared no competing interest.

ICML Workshop

Continuous Vector Quantile Regression

S. Vedula, I. Tallini, A. Rosenberg, and 4 more authors

In ICML Workshop on New Frontiers in Learning, Control, and Dynamical Systems, 2023

Bib

@inproceedings{vedula2023continuous,
  title = {Continuous Vector Quantile Regression},
  author = {Vedula, S. and Tallini, I. and Rosenberg, A. and Pegoraro, M. and Rodola, E. and Romano, Y. and Bronstein, A.},
  booktitle = {ICML Workshop on New Frontiers in Learning, Control, and Dynamical Systems},
  year = {2023},
}

AISTATS

Vector Quantile Regression on Manifolds

M. Pegoraro, S. Vedula, A. Rosenberg, and 3 more authors

In Proc. International Conference on Artificial Intelligence and Statistics (AISTATS), 2024

Bib

@inproceedings{pegoraro2023vector,
  title = {Vector Quantile Regression on Manifolds},
  author = {Pegoraro, M. and Vedula, S. and Rosenberg, A. and Tallini, I. and Rodola, E. and Bronstein, A.},
  booktitle = {Proc. International Conference on Artificial Intelligence and Statistics (AISTATS)},
  year = {2024},
}

ICLR

Fast Nonlinear Vector Quantile Regression

A. Rosenberg, S. Vedula, Y. Romano, and 1 more author

In Proc. International Conference on Learning Representations (ICLR), 2023

Bib HTML

@inproceedings{rosenberg2023fast,
  title = {Fast Nonlinear Vector Quantile Regression},
  author = {Rosenberg, A. and Vedula, S. and Romano, Y. and Bronstein, A.},
  booktitle = {Proc. International Conference on Learning Representations (ICLR)},
  year = {2023},
}

MSML

Spectral Geometric Matrix Completion

A. Boyarski, S. Vedula, and A. Bronstein

In Proceedings of the Second Mathematical and Scientific Machine Learning Conference (MSML), 2021

Bib

@inproceedings{boyarski2021spectral,
  title = {Spectral Geometric Matrix Completion},
  author = {Boyarski, A. and Vedula, S. and Bronstein, A.},
  booktitle = {Proceedings of the Second Mathematical and Scientific Machine Learning Conference (MSML)},
  year = {2021},
}