Elisabetta Carlini
Associate Professor of Numerical Analysis
Department of Mathematics, Sapienza University of Rome
I work on numerical methods for nonlinear partial differential equations: semi-Lagrangian and finite difference schemes for Hamilton–Jacobi–Bellman equations, Mean Field Games, Fokker–Planck equations, problems on networks, and applications to image processing and crowd dynamics.
- Value function u: Hamilton–Jacobi equation, solved backward in time
- Agent density m: Fokker–Planck equation, forward in time
- Semi-Lagrangian step: follow the characteristic from the node and interpolate at its foot
Research
Hamilton–Jacobi–Bellman equations
Semi-Lagrangian and finite difference schemes for first- and second-order HJB equations. High-order schemes coupled with WENO and CWENO interpolation, algorithms in high dimension, generalized Fast Marching methods for fronts with sign-changing and non-local velocities (dislocation dynamics), and the treatment of nonlinear Neumann and Dirichlet boundary conditions.
Mean Field Games
Discretization of the coupled Hamilton–Jacobi / Fokker–Planck system describing equilibria of games with a very large number of agents. Semi-Lagrangian and Lagrange–Galerkin schemes with convergence results, high-order schemes, Newton-type methods, tensor-train schemes for high-dimensional problems, and applications to the Hughes model for crowd evacuation.
Equations on networks
Numerical approximation of Hamilton–Jacobi equations and MFG models on graphs, including the discretization of vertex conditions, convergence analysis and error estimates. Applications to traffic models and front propagation.
Mean curvature motion and image processing
Semi-Lagrangian schemes for mean curvature motion, the AMSS model and area-preserving flows; surface reconstruction from sparse data with radial basis functions; schemes for the game p-Laplacian and ∞-Laplacian with applications to image inpainting.
Publications
- E. Carlini, A. Picarelli, F. J. SilvaA semi-Lagrangian scheme for Hamilton–Jacobi–Bellman equations with Dirichlet boundary conditionsMathematics of Computation in press · arXiv:2503.20283
- 2027E. Carlini, A. ZorkotNewton methods for mean field games: a numerical studyJournal of Computational and Applied Mathematics 490 · arXiv:2512.12752
- 2026E. Carlini, L. SaluzziHigh order tensor-train-based schemes for high-dimensional mean field gamesSIAM Journal on Scientific Computing 48(4), A2027–A2056 · arXiv:2510.15603
- 2026E. Carlini, V. CoscettiA semi-Lagrangian scheme for first-order mean field games based on monotone operatorsIMA Journal of Numerical Analysis · arXiv:2506.10509
- 2025E. Carlini, V. Coscetti, M. PozzaError estimate for a semi-Lagrangian scheme for Hamilton–Jacobi equations on networksCommunications on Applied Mathematics and Computation · DOI · arXiv:2411.02356
- 2025E. Carlini, A. SiconolfiNumerical analysis of time-dependent Hamilton–Jacobi equations on networksNumerische Mathematik 157(6), 2055–2096 · DOI · arXiv:2310.06092
- 2024E. Carlini, R. Ferretti, M. Semplice, S. PredaA CWENO large time-step scheme for Hamilton–Jacobi equationsCommunications on Applied Mathematics and Computation · arXiv:2402.15367
- 2024E. Carlini, S. TozzaA convergent semi-Lagrangian scheme for the game ∞-LaplacianDynamic Games and Applications 15(2), 406–416
- 2024E. Calzola, E. Carlini, F. J. SilvaA high-order scheme for mean field gamesJournal of Computational and Applied Mathematics 445, 115769 · arXiv:2207.08463
- 2024E. Carlini, F. J. Silva, A. ZorkotA Lagrange–Galerkin scheme for first order mean field game systemsSIAM Journal on Numerical Analysis 62(1), 167–198
- 2024E. Carlini, S. TozzaA scheme for the game p-Laplacian and its application to image inpaintingApplied Mathematics and Computation · DOI
- 2023E. Calzola, E. Carlini, X. Dupuis, F. J. SilvaA semi-Lagrangian scheme for Hamilton–Jacobi–Bellman equations with oblique derivatives boundary conditionsNumerische Mathematik 153(1), 49–84 · arXiv:2109.10228
- 2021L. Bonaventura, E. Calzola, E. Carlini, R. FerrettiSecond order fully semi-Lagrangian discretizations of advection–diffusion–reaction systemsJournal of Scientific Computing 88, 1–29 · arXiv:2002.04381
- 2020E. Carlini, A. Festa, N. ForcadelA semi-Lagrangian scheme for Hamilton–Jacobi–Bellman equations on networksSIAM Journal on Numerical Analysis 58, 3165–3196 · arXiv:1804.09429
- 2020L. Bonaventura, E. Calzola, E. Carlini, R. FerrettiA fully semi-Lagrangian method for the Navier–Stokes equations in primitive variables proceedingsLecture Notes in Computational Science and Engineering 132, 55–62 (FEF 2017)
- 2019E. Carlini, M. Falcone, R. FerrettiNumerical techniques for level set models: an image segmentation perspective book chapterIn A. El-Baz, J. S. Suri (eds.), Level Set Method in Medical Imaging Segmentation, Taylor & Francis
- 2018E. Carlini, F. J. SilvaOn the discretization of some nonlinear Fokker–Planck–Kolmogorov equations and applicationsSIAM Journal on Numerical Analysis 56(4), 2148–2177 · arXiv:1708.02042
- 2018E. Carlini, F. Camilli, C. MarchiA flame propagation model on a network with application to a blocking problemDiscrete and Continuous Dynamical Systems – Series S 11(5), 825–843 · arXiv:1411.3260
- 2018E. Carlini, F. J. SilvaA fully-discrete scheme for systems of nonlinear Fokker–Planck–Kolmogorov equations book chapterIn PDE Models for Multi-Agent Phenomena, Springer INdAM Series 28, 195–218 · arXiv:1711.10426
- 2017E. Carlini, A. Festa, F. J. Silva, M.-T. WolframA semi-Lagrangian scheme for a modified version of the Hughes model for pedestrian flowDynamic Games and Applications 7(4), 683–705 · arXiv:1601.07324
- 2017E. Carlini, R. FerrettiA semi-Lagrangian scheme with radial basis approximation for surface reconstructionComputing and Visualization in Science 18(2–3), 103–112 · arXiv:1602.04626
- 2017E. Carlini, A. Festa, F. J. SilvaThe Hughes model for pedestrian dynamics and congestion modelling proceedingsIFAC-PapersOnLine 50(1), 1655–1660 · arXiv:1611.06848
- 2016E. Carlini, F. J. SilvaA semi-Lagrangian scheme for the Fokker–Planck equation proceedingsIFAC-PapersOnLine 49(8), 272–277
- 2015E. Carlini, F. J. SilvaA semi-Lagrangian scheme for a degenerate second order mean field game systemDiscrete and Continuous Dynamical Systems – Series A 35(9), 4269–4292 · arXiv:1404.5932
- 2015E. Carlini, F. Camilli, C. MarchiA model problem for mean field games on networksDiscrete and Continuous Dynamical Systems – Series A 35(9), 4173–4192 · arXiv:1402.2467
- 2014E. Carlini, F. J. SilvaA fully discrete semi-Lagrangian scheme for a first order mean field games problemSIAM Journal on Numerical Analysis 52(1), 45–67 · arXiv:1212.4757
- 2013E. Carlini, R. FerrettiA semi-Lagrangian approximation for the AMSS model of image processingApplied Numerical Mathematics 73, 16–32
- 2013E. Carlini, M. Falcone, P. HochA generalized fast marching method on unstructured triangular meshesSIAM Journal on Numerical Analysis 51(6), 2999–3035
- 2013E. Carlini, F. J. SilvaSemi-Lagrangian schemes for mean field game models proceedings52nd IEEE Conference on Decision and Control (CDC), 3115–3120
- 2013E. Carlini, M. Falcone, A. FestaA brief survey on semi-Lagrangian schemes for image processing book chapterIn Innovations for Shape Analysis, Mathematics and Visualization, Springer, 191–218
- 2012E. Carlini, R. FerrettiA semi-Lagrangian scheme for area preserving flows proceedings21st International Conference on Pattern Recognition (ICPR), Tsukuba, 2029–2033
- 2011E. Carlini, N. Forcadel, R. MonneauA generalized fast marching method for dislocation dynamicsSIAM Journal on Numerical Analysis 49(6), 2470–2500
- 2010E. Carlini, M. Falcone, R. FerrettiConvergence of a large time-step scheme for mean curvature motionInterfaces and Free Boundaries 12(4), 409–441
- 2010A. M. Bersani, E. Carlini, P. Lanucara, M. Rorro, V. RuggieroApplication of optimal control techniques and advanced computing to the study of enzyme kineticsMathematics and Computers in Simulation 81(3), 705–716
- 2008E. Carlini, M. Falcone, N. Forcadel, R. MonneauConvergence of a generalized fast marching method for an eikonal equation with a velocity changing signSIAM Journal on Numerical Analysis 46(6), 2920–2952
- 2008E. Carlini, R. FerrettiA semi-Lagrangian approximation of min-max type for the stationary mean curvature equation proceedingsENUMATH 2007, Springer, 679–686
- 2008E. CarliniA single-pass scheme for the mean curvature motion of convex curves proceedingsENUMATH 2007, Springer, 671–678
- 2007E. Carlini, M. Falcone, R. FerrettiA semi-Lagrangian scheme for the curve shortening flow in codimension 2Journal of Computational Physics 225(2), 1388–1408
- 2006O. Alvarez, E. Carlini, R. Monneau, E. RouyConvergence of a first order scheme for a non-local eikonal equationApplied Numerical Mathematics 56(9), 1136–1146
- 2006O. Alvarez, E. Carlini, R. Monneau, E. RouyA convergent scheme for a non local Hamilton–Jacobi equation modelling dislocation dynamicsNumerische Mathematik 104(4), 413–444
- 2006E. Carlini, E. Cristiani, N. ForcadelA non-monotone fast marching scheme for a Hamilton–Jacobi equation modelling dislocation dynamics proceedingsENUMATH 2005, Springer, 724–731
- 2006E. Carlini, M. Falcone, R. FerrettiA time-adaptive semi-Lagrangian approximation to mean curvature motion proceedingsENUMATH 2005, Springer, 732–739
- 2005O. Alvarez, E. Carlini, P. Hoch, Y. Le Bouar, R. MonneauDislocation dynamics described by non-local Hamilton–Jacobi equationsMaterials Science & Engineering A 400–401, 162–165
- 2005E. Carlini, R. Ferretti, G. RussoA weighted essentially non-oscillatory, large time-step scheme for Hamilton–Jacobi equationsSIAM Journal on Scientific Computing 27(3), 1071–1091
- 2004E. Carlini, M. Falcone, R. FerrettiAn efficient algorithm for Hamilton–Jacobi equations in high dimensionComputing and Visualization in Science 7(1), 15–29
Teaching
Courses in academic year 2026–2027
- MScNumerical Methods for Nonlinear Partial Differential Equations
- MScNumerical Methods for Partial Differential Equations
- BScMathematics with Elements of Computer Science
PhD courses and schools
- 2025Semi-Lagrangian schemes for Hamilton–Jacobi equations
- 2024Approximation of Mean Field Games and related problems
- 2022Approximation of Mean Field Games and related problems
- 2015Numerical Methods for Partial Differential Equations
Previous teaching
- Numerical Analysis (Istituzioni di Analisi Numerica), MSc in Mathematics for Applications, 2020–2026
- Programming and Computing Lab, BSc in Mathematics, 2011–2024
- Mathematical Methods for Engineering (MAT/08 module), Chemical Engineering, 2016–2018
- Computer skills: Matlab and Python, BSc in Mathematics, 2008–2023
- Numerical Analysis and C Programming, Master in Scientific Computing, 2008–2017
- Numerical methods for ordinary differential equations, BSc in Mathematics, 2009–2010
- Numerical Computing 1 and 2, BSc in Chemistry, 2008–2010
- Excellence course “Formulation and numerical approximation of some minimization problems”, with M. Ponsiglione, 2009–2010
- Teaching assistant for Numerical Analysis and Programming and Computing Lab, Sapienza University of Rome and Roma Tre University, 2001–2011
Supervision
PhD students
- 2023–2026Valentina Coscetti · Semi-Lagrangian schemes for first order mean field games and for Hamilton–Jacobi equations on networks
- 2021–2023Ahmad Zorkot · Approximation to Mean Field Games
- 2018–2021Elisa Calzola · Semi-Lagrangian schemes for parabolic equations: second order accuracy and boundary conditions
Theses
I have supervised 22 master's theses in Mathematics for Applications, 18 bachelor's theses, 2 second-level Master's theses in Scientific Computing and 2 international internships (INSA Rouen).
Recent master's theses
- L. Ottavio, Un problema di tipo MFG su networks, 2025–2026
- E. Giansante, Riduzione di modelli per la risoluzione di equazioni di Hamilton–Jacobi, 2024–2025
- G. Testa, Numerical scheme for the Hughes problem using a mean-field game model, 2023–2024
- S. Bonaldo, Algoritmi per il calcolo di equilibri di Nash per giochi con numero finito di giocatori, 2023–2024
- M. Laurenti, Uno schema alle differenze finite per mean-field games, 2023–2024
- A. Mastrodonato, A finite difference approach for solving a mean-field game model of price formation, 2022–2023
- V. Coscetti, Uno schema semi-Lagrangiano per Mean Field Games deterministici basato sulle iterazioni della policy, 2022–2023
- L. Ferentini, Leggi di conservazione su network con schema di Godunov e applicazione al modello LWR, 2021–2022
- F. Puca, Approssimazione numerica del modello di Keller–Segel per la chemiotassi, 2020–2021
- C. Lin, Schema alle differenze finite per equazioni di Hamilton–Jacobi sui network, 2019–2020
- M. Abrugia, Approssimazione numerica di equazioni di Hamilton–Jacobi sui network, 2018–2019
Thesis proposals
Students interested in a thesis on numerical methods for Hamilton–Jacobi equations, Mean Field Games or problems on networks are welcome to write to me to arrange a meeting.
Projects
As principal investigator
- 2025–Mean-Field Games: models, theory, and computational aspects
- 2022Optimal control problems: analysis, approximation and applications
- 2022Numerical approximation of differential models and applications
- 2019Numerical approximation of hyperbolic problems and applications
- 2018International inter-university agreement with IMCA, Lima
- 2013Analysis and approximation of nonlinear differential problems and applications
- 2011Theoretical and applied developments of semi-Lagrangian methods
- 2009Front evolution: models, algorithms and applications
Scientific activities
Recent invited talks
- 2026High order tensor-train-based schemes for high-dimensional mean field games
- 2026Error estimate for a semi-Lagrangian scheme for a time-dependent Hamilton–Jacobi equation on networks and High order tensor-train-based schemes for high-dimensional mean field games
- 2026Algorithm for deterministic mean-field games based on monotone operators
- 2025Numerical approximations of the Newton algorithm for mean field games
- 2025Numerical approximations of the Newton algorithm for mean field games
- 2025Algorithm for deterministic mean-field games
- 2024Algorithm for deterministic mean-field games
- 2024Newton methods for mean field games
- 2023A high-order Lagrange–Galerkin scheme for a class of Fokker–Planck equations and applications to mean field games
- 2023Lagrange–Galerkin schemes for mean field games
- 2023Lagrange–Galerkin schemes for mean field games
- 2023A high-order scheme for mean field games
- 2023A high-order Lagrange–Galerkin scheme for mean field games
- 2023A numerical scheme for evolutive Hamilton–Jacobi equations on networks
- 2023A numerical scheme for evolutive Hamilton–Jacobi equations on networks
- 2023A numerical scheme for evolutive Hamilton–Jacobi equations on networks
Conference organization
Other conferences organized (2011–2023)
- Maurizio day: numerics for controls, PDEs and images, Sapienza University of Rome, 2023
- Nonlinear hyperbolic PDEs: numerical techniques and related models, minisymposium at WONAPDE, Concepción, Chile
- Numerical approximation of hyperbolic problems and applications, GNCS Project, Sapienza University of Rome, 2020
- Numerical methods for optimal control problems and differential games, minisymposium at SIMAI, Milan, 2016
- Numerical methods for PDEs: optimal control, games and image processing, Sapienza University of Rome, 2014
- Numerics in Image and Surface Processing, Bologna, 2014
- Mathematical methods in image processing, Sapienza University of Rome, 2013
- Recent advances on theory and applications of semi-Lagrangian methods, Sapienza University of Rome, 2011
Referee activity
SIAM Journal on Numerical Analysis, SIAM Journal on Control and Optimization, SIAM Journal on Scientific Computing, Mathematics of Computation, ESAIM: M2AN, Mathematical Methods in the Applied Sciences, Computers & Mathematics with Applications, Applied Mathematics and Computation, Dynamic Games and Applications, IEEE Transactions on Automatic Control, IEEE Conference on Decision and Control, Communications in Mathematical Sciences, IEEE Journal of Selected Topics in Signal Processing.
Committees and service
- Examiner and referee of PhD theses at Université Côte d'Azur and Mohammed VI Polytechnic University (2026), University of Tours (2025), University of Rouen (2025), Université de Technologie de Compiègne (2021), GSSI (2020) and Sapienza University of Rome
- Evaluation committees for researcher and associate professor positions in Numerical Analysis (Ferrara, Catania, Verona, Insubria)
- PhD committee in Mathematics, Sapienza University of Rome (2020–2022)
- Thesis committee (2012–2022), library committee, master's study plans committee
- Mathematical Olympiad project, Rome section
Curriculum vitae
- 2026National Scientific Qualification as Full Professor
- 2018–presentAssociate Professor
- 2008–2018Researcher
- 2005–2007Research fellow
- 2004Postdoctoral researcher
- 2004PhD in Mathematical Models and Methods for Technology and Society
- 2002Mathematics in Nanoscale Science and Engineering program
- 2000Degree in Mathematics
Contact
AddressDepartment of Mathematics
Sapienza University of Rome
Piazzale Aldo Moro 5, 00185 Rome, Italy