Publications

2026

The Future of Artificial Intelligence and the Mathematical and Physical Sciences (AI+MPS)

Andrew Ferguson, Marisa LaFleur, Lars Ruthotto, Jesse Thaler, Yuan-Sen Ting, Pratyush Tiwary, Soledad Villar

Machine Learning: Science and Technology 7(2), 023001 [cs.LG]

DOI arXiv

2025

Efficient Neural Network Approaches for Conditional Optimal Transport with Applications in Bayesian Inference

Zheyu Oliver Wang, R. Baptista, Youssef M. Marzouk, Lars Ruthotto, Deepanshu Verma

SIAM Journal on Scientific Computing

DOI Code S2

2024

Multilevel Diffusion: Infinite Dimensional Score-Based Diffusion Models for Image Generation

Paul Hagemann, Lars Ruthotto, G. Steidl, Nicole Yang

SIAM Journal on Mathematics of Data Science

DOI arXiv Code S2

2024

Differential Equations for Continuous-Time Deep Learning

Lars Ruthotto

Notices of the American Mathematical Society

DOI arXiv S2

2024

Neural network approaches for parameterized optimal control

Deepanshu Verma, Nickolas Winovich, Lars Ruthotto, B. V. B. Waanders

Foundations of Data Science

DOI arXiv S2

2024

PyHySCO: GPU-enabled susceptibility artifact distortion correction in seconds

Abigail Julian, L. Ruthotto

Frontiers in Neuroscience

DOI arXiv S2

2024

ACID: A comprehensive toolbox for image processing and modeling of brain, spinal cord, and ex vivo diffusion MRI data

Gergely Dávid, Björn Fricke, J. Oeschger, L. Ruthotto, Francisco J. Fritz, Ora Ohana, L. Mordhorst, Thomas Sauvigny, Patrick Freund, K. Tabelow, S. Mohammadi

bioRxiv

DOI S2

2023

Alternating Minimization for Regression with Tropical Rational Functions

Alex Dunbar, Lars Ruthotto

Algebraic Statistics

DOI arXiv S2

2023

LSEMINK: a modified Newton–Krylov method for Log-Sum-Exp minimization

Kan K, Nagy JG, Lars Ruthotto

Electronic Transactions on Numerical Analysis 618-635

DOI arXiv S2

2022

A Neural Network Approach for Stochastic Optimal Control

Xingjian Li, Deepanshu Verma, Lars Ruthotto

SIAM Journal on Scientific Computing 46(5), C535-C556

DOI arXiv S2

2022

'hessQuik': Fast Hessian computation of composite functions

Elizabeth Newman, Lars Ruthotto

Journal of Open Source Software

DOI Code S2

2021

An introduction to deep generative modeling

Lars Ruthotto, E. Haber

GAMM-Mitteilungen

DOI arXiv Code S2

2021

A Neural Network Approach for High-Dimensional Optimal Control Applied to Multiagent Path Finding

Derek Onken, L. Nurbekyan, Xingjian Li, Wu Fung S, S. Osher, Lars Ruthotto

IEEE Transactions on Control Systems Technology

DOI arXiv Code S2

2021

slimTrain - A Stochastic Approximation Method for Training Separable Deep Neural Networks

Elizabeth Newman, Julianne Chung, Matthias Chung, Lars Ruthotto

SIAM Journal on Scientific Computing 44(4), 1419-1457

DOI arXiv S2

2020

PNKH-B: A Projected Newton-Krylov Method for Large-Scale Bound-Constrained Optimization

Kan K, Wu Fung S, Lars Ruthotto

SIAM Journal on Scientific Computing

DOI arXiv S2

2020

Train Like a (Var)Pro: Efficient Training of Neural Networks with Variable Projection

Elizabeth Newman, Lars Ruthotto, Joseph L. Hart, B. V. B. Waanders

SIAM Journal on Mathematics of Data Science 3(4)

DOI arXiv S2

2020

MGIC: Multigrid-in-Channels Neural Network Architectures

Moshe Eliasof, Jonathan Ephrath, Lars Ruthotto, Eran Treister

SIAM Journal on Scientific Computing S307-S328

DOI arXiv S2

2020

Inversion of convection–diffusion equation with discrete sources

Meenarli Sharma, Mirko Hahn, Sven, Leyffer, Lars Ruthotto, Bart van, Bloemen Waanders

Optimization and Engineering

DOI S2

2019

LeanConvNets: Low-Cost Yet Effective Convolutional Neural Networks

Jonathan Ephrath, Moshe Eliasof, Lars Ruthotto, E. Haber, Eran Treister

IEEE Journal on Selected Topics in Signal Processing 14(4), 894-904

DOI arXiv S2

2019

A machine learning framework for solving high-dimensional mean field game and mean field control problems

Lars Ruthotto, S. Osher, Wuchen Li, L. Nurbekyan, Wu Fung S

Proceedings of the National Academy of Sciences of the United States of America 117(17), 9183-9193

DOI arXiv Code S2

2019

hMRI – A toolbox for quantitative MRI in neuroscience and clinical research

K. Tabelow, E. Balteau, J. Ashburner, M. Callaghan, Bogdan Draganski, G. Helms, Ferath Kherif, T. Leutritz, A. Lutti, C. Phillips, Enrico Reimer, Lars Ruthotto, Maryam Seif, N. Weiskopf, G. Ziegler, S. Mohammadi

NeuroImage

DOI S2

2019

ADMM-Softmax: an ADMM approach for multinomial logistic regression

Wu Fung S, Sanna Tyrväinen, Lars Ruthotto, E. Haber

Electronic Transactions on Numerical Analysis 214-229

DOI S2

2018

LAP: A Linearize and Project Method for Solving Inverse Problems with Coupled Variables

James L. Herring, Nagy JG, Lars Ruthotto

Sampling Theory, Signal Processing, and Data Analysis

DOI arXiv

2018

A Bayesian framework for molecular strain identification from mixed diagnostic samples

Lauri Mustonen, Xiang-hui Gao, Asteroide Santana, Rebecca Mitchell, Ymir Vigfusson, Lars Ruthotto

Inverse Problems

DOI arXiv S2

2018

Deep Neural Networks Motivated by Partial Differential Equations

Lars Ruthotto, E. Haber

Journal of Mathematical Imaging and Vision

DOI arXiv S2

2018

An Uncertainty-Weighted Asynchronous ADMM Method for Parallel PDE Parameter Estimation

Wu Fung S, Lars Ruthotto

SIAM Journal on Scientific Computing 41(5)

DOI arXiv S2

2018

Layer-Parallel Training of Deep Residual Neural Networks

Stefanie Günther, Lars Ruthotto, J. Schroder, E. Cyr, N. Gauger

SIAM Journal on Mathematics of Data Science 2(1), 1-23

DOI arXiv S2

2018

Gauss–Newton Optimization for Phase Recovery From the Bispectrum

James Herring, Nagy JG, Lars Ruthotto

IEEE Transactions on Computational Imaging 235-247

DOI arXiv S2

2018

Numerical methods for polyline‐to‐point‐cloud registration with applications to patient‐specific stent reconstruction

C. Y. Lin, A. Veneziani, Lars Ruthotto

International Journal for Numerical Methods in Biomedical Engineering 34(3)

DOI S2

2018

Improved Susceptibility Artifact Correction of Echo-Planar MRI using the Alternating Direction Method of Multipliers

J. Macdonald, Lars Ruthotto

Journal of Mathematical Imaging and Vision

DOI S2

2018

Optimal Experimental Design for Inverse Problems with State Constraints

Lars Ruthotto, Julianne Chung, Matthias Chung

SIAM Journal on Scientific Computing 40(4), B1080-B1100

DOI Code S2

2017

A Lagrangian Gauss-Newton-Krylov Solver for Mass- and Intensity-Preserving Diffeomorphic Image Registration

A. Mang, Lars Ruthotto

SIAM Journal on Scientific Computing 39(5), B5860-B5885

DOI arXiv S2

2017

Stable architectures for deep neural networks

E. Haber, Lars Ruthotto

Inverse Problems

DOI arXiv Code S2

2017

A multiscale method for model order reduction in PDE parameter estimation

Wu Fung S, Lars Ruthotto

Journal of Computational and Applied Mathematics 19-34

DOI arXiv S2

2017

Computational methods for image reconstruction

Julianne Chung, Lars Ruthotto

NMR in Biomedicine

DOI Code S2

2017

A stabilized multigrid solver for hyperelastic image registration

Lars Ruthotto, C. Greif, J. Modersitzki

Numerical Linear Algebra with Applications

DOI S2

2016

jInv-a Flexible Julia Package for PDE Parameter Estimation

Lars Ruthotto, Eran Treister, E. Haber

SIAM Journal on Scientific Computing 39(5), S702-S722

DOI arXiv Code S2

2016

Efficient Numerical Optimization For Susceptibility Artifact Correction Of EPI-MRI

J. Macdonald, L. Ruthotto

arXiv.org

arXiv S2

2015

High-resolution diffusion kurtosis imaging at 3T enabled by advanced post-processing

S. Mohammadi, K. Tabelow, L. Ruthotto, T. Feiweier, J. Polzehl, N. Weiskopf

Frontiers in Neuroscience

DOI S2

2014

A multiscale finite volume method for Maxwell's equations at low frequencies

E. Haber, L. Ruthotto

Geophysical Journal International 199(2), 1268-1277

DOI S2

2014

Geophysical Imaging of Fluid Flow in Porous Media

J. Fohring, E. Haber, Lars Ruthotto

SIAM Journal on Scientific Computing 36(5), S218-S236

DOI S2

2013

A Hyperelastic Regularization Energy for Image Registration

M. Burger, J. Modersitzki, Lars Ruthotto

SIAM Journal on Scientific Computing 35(1), B132-B148

DOI S2

2012

Diffeomorphic susceptibility artifact correction of diffusion-weighted magnetic resonance images

L. Ruthotto, L. Ruthotto, H. Kugel, J. Olesch, B. Fischer, J. Modersitzki, M. Burger, C. Wolters

Physics in Medicine and Biology

DOI S2

2012

Motion Correction in Dual Gated Cardiac PET Using Mass-Preserving Image Registration

F. Gigengack, Lars Ruthotto, M. Burger, C. Wolters, Xiaoyi Jiang, K. Schäfers

IEEE Transactions on Medical Imaging 31(3), 698-712

DOI S2

2012

Sensitivity of EEG leads to volume conductor properties

S. Wagner, J. Vorwerk, L. Ruthotto, H. Kugel, M. Burger, T. Knösche, B. Maess, Carsten H. Wolters

Biomedizinische Technik. Biomedical engineering

DOI S2

2026

Manifold-Aware Perturbations for Constrained Generative Modeling

Katie Keegan, Lars Ruthotto

International Conference on Machine Learning (Spotlight) [cs.LG]

arXiv

2026

RAPNet: Accelerating Algebraic Multigrid with Learned Sparse Corrections

Yali Fink, Ido Ben-Yair, Lars Ruthotto, Eran Treister

International Conference on Machine Learning

2023

Learning Control Policies of Hodgkin-Huxley Neuronal Dynamics

Malvern Madondo, Deepanshu Verma, Lars Ruthotto, N. A. Yong

arXiv.org

DOI arXiv S2

2022

Multivariate Quantile Function Forecaster

Kan K, Franccois-Xavier Aubet, Tim Januschowski, Youngsuk Park, Konstantinos Benidis, Lars Ruthotto, Jan Gasthaus

International Conference on Artificial Intelligence and Statistics

arXiv S2

2022

Improving Graph Neural Networks with Learnable Propagation Operators

Moshe Eliasof, Lars Ruthotto, Eran Treister

International Conference on Machine Learning

arXiv S2

2022

ACID - an open-source, bids compatible software for brain and spinal cord dMRI: preprocessing, DTI/DKI, biophysical modelling

Björn Fricke, Gergely Dávid, J. Oeschger, P. Freund, L. Ruthotto, K. Tabelow, S. Mohammadi

ISMRM Annual Meeting

DOI S2

2021

Multigrid-in-Channels Architectures for Wide Convolutional Neural Networks

Moshe Eliasof, Jonathan Ephrath, Lars Ruthotto, Eran Treister

Copper Mountain Conference on Multigrid Methods; NeurIPS 2021 Workshop on Deep Learning and PDEs

arXiv

2020

OT-Flow: Fast and Accurate Continuous Normalizing Flows via Optimal Transport

Derek Onken, Wu Fung S, Xingjian Li, Lars Ruthotto

AAAI Conference on Artificial Intelligence

DOI arXiv Code S2

2020

A Neural Network Approach Applied to Multi-Agent Optimal Control

Derek Onken, L. Nurbekyan, Xingjian Li, Wu Fung S, S. Osher, Lars Ruthotto

European Control Conference

DOI arXiv S2

2019

IMEXnet: A Forward Stable Deep Neural Network

E. Haber, Keegan Lensink, Eran Treister, Lars Ruthotto

International Conference on Machine Learning

arXiv S2

2019

LeanResNet: A Low-cost yet Effective Convolutional Residual Networks

Jonathan Ephrath, Lars Ruthotto, E. Haber, Eran Treister

International Conference on Machine Learning

arXiv S2

2018

Learning Across Scales - Multiscale Methods for Convolution Neural Networks

E. Haber, Lars Ruthotto, E. Holtham

AAAI Conference on Artificial Intelligence

DOI S2

2017

Reversible Architectures for Arbitrarily Deep Residual Neural Networks

B. Chang, Lili Meng, E. Haber, Lars Ruthotto, David Begert, E. Holtham

AAAI Conference on Artificial Intelligence

DOI arXiv S2

2016

Combined Background Field Removal and Reconstruction for Quantitative Susceptibility Mapping

M. März, Lars Ruthotto

Bildverarbeitung für die Medizin

DOI S2

2014

Model-Based Parameterestimation in DCE-MRI Without an Arterial Input Function

C. Heck, Lars Ruthotto, J. Modersitzki, B. Berkels

Bildverarbeitung für die Medizin

DOI S2

2014

A new method for joint susceptibility artefact correction and super-resolution for dMRI

Lars Ruthotto, S. Mohammadi, N. Weiskopf

Medical Imaging

DOI S2

2014

A Multiscale Method for Fast Electromagnetic Simulations

E. Haber, L. Ruthotto

SEG Technical Program Expanded Abstracts

DOI S2

2013

HySCO - Hyperelastic Susceptibility Artefact Correction of DTI in SPM

Lars Ruthotto, Siawoosh Mohammadi, Constantin Heck, Jan Modersitzki, Nikolaus Weiskopf

Bildverarbeitung für die Medizin 344-349

DOI

2013

Geophysical imaging, reservoir history matching and forecasting

J. Fohring, L. Ruthotto, E. Haber

SEG Technical Program Expanded Abstracts

DOI S2

2012

A Simplified Pipeline for Motion Correction in Dual Gated Cardiac PET

Lars Ruthotto, F. Gigengack, M. Burger, C. Wolters, Xiaoyi Jiang, K. Schäfers, J. Modersitzki

Bildverarbeitung für die Medizin 51-56

DOI S2

2012

Registration of Dynamic Contrast Enhanced MRI with Local Rigidity Constraint

Lars Ruthotto, E. Hodneland, J. Modersitzki

Workshop on Biomedical Image Registration 190-198

DOI S2

2012

Atlas-Based Whole-Body PET-CT Segmentation Using a Passive Contour Distance

F. Gigengack, Lars Ruthotto, Xiaoyi Jiang, J. Modersitzki, M. Burger, S. Hermann, K. Schäfers

MCV

DOI S2

2012

Pipeline for motion correction in dual gated PET

F. Gigengack, L. Ruthotto, T. Kosters, Xiaoyi Jiang, J. Modersitzki, M. Burger, C. Wolters, K. Schafers

Nuclear Science Symposium and Medical Imaging Conference

DOI S2

2012

Atlas-based segmentation using passive contours

F. Gigengack, L. Ruthotto, T. Kosters, Xiaoyi Jiang, J. Modersitzki, M. Burger, S. Hermann, K. Schafers

Nuclear Science Symposium and Medical Imaging Conference

DOI S2

2011

Mass-preserving motion correction of PET: Displacement field vs. spline transformation

F. Gigengack, L. Ruthotto, M. Burger, C. Wolters, Xiaoyi Jiang, K. Schafers

IEEE Nuclear Science Symposium Conference Record

DOI S2

2011

Mass-preserving motion correction of dual gated cardiac PET

F. Gigengack, L. Ruthotto, M. Burger, C. Wolters, Xiaoyi Jiang, K. Schafers

IEEE Nuclear Science Symposium Conference Record

DOI S2

2010

Motion correction of cardiac PET using mass-preserving registration

F. Gigengack, L. Ruthotto, M. Burger, C. Wolters, Xiaoyi Jiang, K. Schafers

IEEE Nuclear Science Symposuim & Medical Imaging Conference 3317-3319

DOI S2

2010

A variational approach for the correction of field-inhomogeneities in EPI sequences

J. Olesch, Lars Ruthotto, H. Kugel, S. Skare, B. Fischer, C. Wolters

Medical Imaging

DOI S2

2015

Non-linear Image Registration

Lars Ruthotto, J. Modersitzki

Handbook of Mathematical Methods in Imaging 2005-2051

DOI S2

2026

Differentiating through Stochastic Differential Equations: A Primer

Rishi Leburu, Levon Nurbekyan, Lars Ruthotto

arXiv.org [math.NA]

arXiv

2025

Weight-Parameterization in Continuous Time Deep Neural Networks for Surrogate Modeling

Haley Rosso, Lars Ruthotto, K. Sargsyan

arXiv.org

DOI arXiv S2

2025

Mixed Precision Training of Neural ODEs

Elena Celledoni, Brynjulf Owren, Lars Ruthotto, Nicole Yang

arXiv.org

arXiv

2020

Discretize-Optimize vs. Optimize-Discretize for Time-Series Regression and Continuous Normalizing Flows

Derek Onken, Lars Ruthotto

arXiv.org

arXiv Code S2

2020

Avoiding The Double Descent Phenomenon of Random Feature Models Using Hybrid Regularization

Kan K, Nagy JG, Lars Ruthotto

arXiv.org

arXiv S2

2018

Low-Cost Parameterizations of Deep Convolution Neural Networks

Eran Treister, Lars Ruthotto, Moshe Sharoni, Shira Zafrani, Eldad Haber

arXiv.org

arXiv

2018

Never Look Back - A Modified EnKF Method and its Application to the Training of Neural Networks without Back Propagation

Eldad Haber, Felix Lucka, Lars Ruthotto

arXiv.org

arXiv