Papers
Papers
- Neumann eigenmaps for landmark embedding (2025).Shashank Sule, Wojciech Czaja. [arXiv]
- Input layer regularization and automated regularization hyperparameter tuning for myelin water estimation using deep learning (2025). Mirage Modi, Shashank Sule, Jonathan Palumbo, Michael Rozowski, Mustapha Bouhrara, Wojciech Czaja, Richard G. Spencer. [arXiv]
- Neural network-based speech reconstruction from undersampled STFT magnitude data. Wojciech Czaja, Matthias Wellershoff, Shashank Sule, Canran Ji. In the 32nd European Signal Processing Conference 2024.
- On the limits of neural network explainability via descrambling (2024). Shashank Sule, Richard G. Spencer, Wojciech Czaja [arXiv] (Submitted, in review).
- Sharp error estimates for target measure diffusion maps and applications to the committor problem (2023). Shashank Sule, Luke Evans, Maria Cameron [arXiv] (Submitted, in review).
- Sobolev Orthogonal Polynomials on the Sierpinski Gasket. Qingxuan Jiang, Tian Lan, Kasso Okoudjou, Robert Strichartz, Shashank Sule, and Sreeram Venkat. , J Fourier Anal Appl 27, 38 (2021). [arXiv] [code]
Selected Talks
-
Learning collective variables for quantifiably good representations of molecular dynamics, Learning Collective Variables and Coarse Grained Models, Institute of Statistical and Mathematical Innovation, UChicago.
-
SOPs on the Sierpinski Gasket, AMS Contributed Paper Session on Functional Analysis, Operator Theory, and Operator Algebras 1, Joint Mathematics Meetings 2020.
Selected Poster Presentations
-
Error analysis of Target Measure Diffusion Maps, Measure Transport, Diffusion Processes and Sampling Workshop, Flatiron Institute.
-
Neural network interpretation using descrambler groups: do all roads lead to the SVD?, Fall Fourier Talks 2022