My current research lies broadly in geometric statistics and learning on singular spaces. I'm also interested in concentration of measure and random matrix theory. Here are links my pages on ResearchGate, Google Scholar, where most of my publications and available online.
I'm very much open to academic research collaboration in the areas that interest me or have significant overlaps with them. So please feel free to contact me for the same, and I'd be pleased to have a discussion!
(P: pure mathematics, A: applied mathematics, SP: statistics/probability, ML: machine learning)
A Fanning Scheme for the Parallel Transport Along Geodesics on Riemannian Manifolds. Susovan Pal, Maxime Louis, Benjamin Charlier, Paul Jusselin, Stanley Durrleman. 2018. SIAM Journal on Numerical Analysis, 56(4):2563–2584. DOI: 10.1137/17M1130617. Link. (P, A).
A Riemannian Framework for Linear and Quadratic Discriminant Analysis on the Tangent Space of Shapes.
Susovan Pal, Roger P. Woods, Shantanu Joshi, Suchit Panjoyar, Elizabeth Sowel, Katherine L. Narr. 2017. IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). DOI: 10.1109/CVPRW.2017.102 — Best paper award. Link. (ML) ArXiV. (SP).
Construction of a closed hyperbolic surface of arbitrarily small eigenvalue of prescribed serial number. Susovan Pal, Contemporary Mathematics, Vol. 590, 2012. Link. (P)
Splintering with distributions and polytopes: Unconventional schemes for private computation.
Praneeth Vepakomma, Julia Balla, Susovan Pal, Ramesh Raskar. 2021. IEEE EMBS Grand Challenges Forum on COVID-19 Tracing, virtual event. Link. (ML)
Accepted subject to minor revision / In Press
Convergence of Mean Shift Algorithms for large bandwidths and simultaneous efficient clustering.
Susovan Pal, Praneeth Vepakomma. 2025. International Journal of Data Science and Analytics (Springer). Conditionally accepted (minor revision). ArXiV preprint.. Link to talk. (P, ML)
Manifolds with Non-Smooth Boundaries and Asymptotics of the Graph Laplacian. Susovan Pal, David Tewodrose. Proceedings of Machine Learning Research (PMLR), GTML 2025 , Leipzig, Germany. 2025. Conditionally accepted (minor revision). (ML).
Boundary differentiability of Douady-Earle extensions of diffeomorphisms of Sn. Jun Hu, Susovan Pal, Pure and Applied Mathematics Quarterly, 2013. (P) (revision pending)
Preprints/Submitted:
Manifolds with kinks and the asymptotic behavior of the graph Laplacian operator with Gaussian kernel. Susovan Pal, David Tewodrose. 2025. Submitted. ArXiV preprint. (P, SP, ML).
Asymptotics of the Graph Laplace Operator near an isolated singularity. 2025. Susovan Pal. Submitted. ArXiV preprint. (P, ML).
Two-Sample Tests for Optimal Lifts, Manifold Stability and Reverse Labeling Reflection Shape. 2025. Do Tran Van, Susovan Pal, Benjamin Eltzner, Stephan Huckemann. ArXiv preprint. Link to talk. (P, SP).
Identifiability of the Unnormalized Graph Laplace Operators. 2026. Susovan Pal. Submitted. ArXiV preprint.
In preparation:
Support of Continuous Smeary Measures on Spheres. 2026. Susovan Pal. In preparation. Draft shareable upon request. (SP)
L^p convergence of Asymptotic Mean Value Laplacians on manifolds with singularities. Susovan Pal, Manuel Dias. 2026. Draft shareable upon request. (P)
A generalization of Smeary Central Limit Theorem. 2025.
Do Tran Van, Susovan Pal, Benjamin Eltzner, Stephan Huckemann. Draft shareable upon request. (SP)