Research Output

Publications

Peer-reviewed papers and invited talks on biophysical modeling, statistical inference, and scientific machine learning.

Papers

"Learning biophysical models of gene regulatory dynamics using probability flow matching"
Maddu, S., Chardes, V., Shelley, M. · Under review · 2026
Biophysical ModelingGenerative ModelingInverse Problems
PDF ↗
"Human Mitotic Spindles as Active Liquid Crystals: From Collective Behaviors to Discrete Filaments"
Maddu, S., Kelleher, C. · Proceedings of the National Academy of Sciences (PNAS) · 2026
Cell BiophysicsActive MatterLiquid Crystal Theory
PDF ↗
"Inferring stochastic dynamics with growth from cross-sectional data"
Zhang, S., Maddu, S., Qiu, X., Chardes, V. · Neural Information Processing Systems (NeurIPS) · 2025
Stochastic InferenceSingle-Cell DataGenerative Modeling
Paper ↗
"Inferring biological processes with intrinsic noise from cross-sectional data"
Maddu, S., Chardes, V. · Proceedings of the National Academy of Sciences (PNAS) · 2025
Stochastic InferenceSingle-Cell DataGenerative Modeling
PDF ↗
"The Well: a Large-Scale Collection of Diverse Physics Simulations for Machine Learning"
Datasets and Benchmarks Track, NeurIPS · 2024 (Maddu, S., domain expert contributor)
Physics SimulationsMachine LearningDatasets & Benchmarks
Paper ↗
"Learning locally dominant force balances in active particle systems"
Sturm, D., Maddu, S., Sbalzarini, I.F. · Proceedings of the Royal Society A · 2024
Active MatterSparse InferenceParticle Systems
Paper ↗
"Learning fast, accurate, and stable closures of a kinetic theory of an active fluid"
Maddu, S. · Journal of Computational Physics · 2024
Neural OperatorsActive FluidsClosure Models
PDF ↗
"Stochastic force inference via density estimation"
Chardes, V., Maddu, S. · NeurIPS 2023 AI for Science Workshop
Stochastic InferenceDensity EstimationSingle-Cell Data
"Adaptive weighting of Bayesian physics informed neural networks for multitask and multiscale forward and inverse problems"
Perez, S., Maddu, S., Sbalzarini, I.F., Poncet, P. · Journal of Computational Physics · 2023
Physics-Informed MLBayesian InferenceMultiscale Modeling
Paper ↗
"STENCIL-NET for equation-free forecasting from data"
Maddu, S. · Nature Scientific Reports · vol. 13, no. 1, 2023
Neural OperatorsEquation-Free ForecastingScientific ML
PDF ↗
"Stability selection enables robust learning of partial differential equations from limited noisy data"
Maddu, S. · Proceedings of the Royal Society A · vol. 478, no. 2262, 2022
PDE InferenceSparse OptimizationStatistical Learning
PDF ↗
"Parallel discrete convolutions on adaptive particle representations of images"
Jonsson, J., Cheeseman, B., Maddu, S., Gonciarz, K., Sbalzarini, I.F. · IEEE Transactions on Image Processing · vol. 31, 2022
Image ProcessingScientific ComputingMicroscopy
Paper ↗
"Inverse-Dirichlet weighting enables reliable training of physics informed neural networks"
Maddu, S. · Machine Learning: Science and Technology · vol. 3, no. 1, 2021
Physics-Informed MLMulti-Objective OptimizationNeural Networks
PDF ↗
"Learning physically consistent differential equation models from data using group sparsity"
Maddu, S. · Physical Review E · vol. 103, no. 4, 2021
PDE InferenceSparse OptimizationGroup Sparsity
PDF ↗

Talks

"Inference of Multiscale Dynamical Systems from Biological Data"
NITMB Workshop — Learning Multivariate Spatio-Temporal Dynamical Models, Chicago, IL
Mar. 2027
"Scalable inference of biophysical models from single-cell data"
Stanford University, Stanford, CA
Aug. 2026
"Scalable inference of biophysical models from single-cell data"
Gladstone Institute, San Francisco, CA
Aug. 2026
"Active Liquid Crystal Theory of Microtubule Organization in Human Spindles"
Theory of Living Matter Seminar
Feb. 2026
"Inference of Multiscale Dynamical Systems from Biological Data"
Soft Matter Seminar, Department of Physics, Syracuse University
Feb. 2026
"Inference of Multiscale Dynamical Systems from Biological Data"
AMS Special Seminar Series, Johns Hopkins University
Jan. 2026
"Active Liquid Crystal Theory of Microtubule Organization in Human Spindles"
Physics of Life Symposium, CZ Biohub SF / UCSF, San Francisco, CA
Sep. 2025
"Scalable Inference of Biophysical Models from Multi-Omics Data"
AI in Molecular Biology, Santa Fe, New Mexico
Sep. 2025
"Scalable inference of biological processes from cross-sectional datasets"
Frontiers in Applied & Computational Mathematics, NJIT
Jun. 2025
"Inferring biological processes with intrinsic noise from cross-sectional data"
Workshop on Modeling and Inference of Stochastic Processes, Flatiron Institute
Nov. 2024
"Integrating mechanistic models with single-cell data via probability flow inference"
Flatiron-wide Algorithms and Mathematics, New York
Oct. 2024
"Probing the physical basis of human mitotic spindle self-organization using data from electron microscopy"
X International Conference on Computational Bioengineering (ICCB), Vienna
Sep. 2023
"Learning accurate closures of a kinetic theory of an active fluid"
APS March Meeting, Las Vegas
Mar. 2023
"Learning solution adaptive discretization stencils from data"
Conference on Parallel Processing for Scientific Computing (SIAM-PP), Seattle
Feb. 2022
"Encoding knowledge with group sparsity for model learning from limited and noisy biological data"
World Congress in Computational Mechanics and ECCOMAS Congress, Paris
Jan. 2021
"Enabling data-driven modeling in biology by statistical learning of interpretable mathematical models from microscopy data"
New York Scientific Data Summit (NYSDS), Columbia University, New York
Jun. 2019
"Enabling data-driven modeling in biology"
Center for Computational Mathematics, Flatiron Institute, New York
Jun. 2019