Stefania Fresca
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Stefania Fresca has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Andrea Manzoni
Politecnico di Milano
15 shared publications - Alejandro Federico Frangi
University of Manchester
4 shared publications - Giorgio Gobat
Politecnico di Milano
4 shared publications - Simone Brivio
Department of Mathematics MOX Politecnico di Milano Milan Italy
4 shared publications - Nicola Rares Franco
MOX, Department of Mathematics, Politecnico di Milano, Milan, Italy
3 shared publications Andrea Manzoni
2 shared publications
Latest Publications
- Open Access
- 30 October 2026
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 464
HypeRL: Hypernetwork-based reinforcement learning for control of parametrized dynamical systems
- 30 October 2026
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 464
Open AccessHypeRL: Hypernetwork-based reinforcement learning for control of parametrized dynamical systems
- Open Access
- 10 February 2026
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 453
Multi-Level Monte Carlo training of neural operators
- 10 February 2026
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 453
Open AccessMulti-Level Monte Carlo training of neural operators
- Open Access
- 21 April 2025
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 442
Handling geometrical variability in nonlinear reduced order modeling through Continuous Geometry-Aware DL-ROMs
- 21 April 2025
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 442
Open AccessHandling geometrical variability in nonlinear reduced order modeling through Continuous Geometry-Aware DL-ROMs
- Open Access
Beyond Interpolation: Extrapolative Reasoning with Reinforcement Learning and Graph Neural Networks
Despite incredible progress, many neural architectures fail to properly generalize beyond their training distribution. As such, learning to reason in a correct and generalizable way is one of the current fundamental challenges in machine learning. In this respect, logic puzzles provide a great testbed, as we can fully understand and control the learning environment. Thus, they allow to evaluate performance on previously unseen, larger and more difficult puzzles that follow the same underlying rules. Since traditional approaches often struggle to represent such scalable logical structures, we propose to model these puzzles using a graph-based approach. Then, we investigate the key factors enabling the proposed models to learn generalizable solutions in a reinforcement learning setting. Our study focuses on the impact of the inductive bias of the architecture, different reward systems and the role of recurrent modeling in enabling sequential reasoning. Through extensive experiments, we demonstrate how these elements contribute to successful extrapolation on increasingly complex puzzles.These insights and frameworks offer a systematic way to design learning-based systems capable of generalizable reasoning beyond interpolation.
Open AccessBeyond Interpolation: Extrapolative Reasoning with Reinforcement Learning and Graph Neural Networks
Despite incredible progress, many neural architectures fail to properly generalize beyond their training distribution. As such, learning to reason in a correct and generalizable way is one of the current fundamental challenges in machine learning. In this respect, logic puzzles provide a great testbed, as we can fully understand and control the learning environment. Thus, they allow to evaluate performance on previously unseen, larger and more difficult puzzles that follow the same underlying rules. Since traditional approaches often struggle to represent such scalable logical structures, we propose to model these puzzles using a graph-based approach. Then, we investigate the key factors enabling the proposed models to learn generalizable solutions in a reinforcement learning setting. Our study focuses on the impact of the inductive bias of the architecture, different reward systems and the role of recurrent modeling in enabling sequential reasoning. Through extensive experiments, we demonstrate how these elements contribute to successful extrapolation on increasingly complex puzzles.These insights and frameworks offer a systematic way to design learning-based systems capable of generalizable reasoning beyond interpolation.
- Open Access
- 17 January 2025
- Published by Elsevier in Neural Networks
- Vol. 185, 107146
On latent dynamics learning in nonlinear reduced order modeling
- 17 January 2025
- Published by Elsevier in Neural Networks
- Vol. 185, 107146
Open AccessOn latent dynamics learning in nonlinear reduced order modeling
- Open Access
Handling geometrical variability in nonlinear reduced order modeling through Continuous Geometry-Aware DL-ROMs
Deep Learning-based Reduced Order Models (DL-ROMs) provide nowadays a well-established class of accurate surrogate models for complex physical systems described by parametrized PDEs, by nonlinearly compressing the solution manifold into a handful of latent coordinates. Until now, design and application of DL-ROMs mainly focused on physically parameterized problems. Within this work, we provide a novel extension of these architectures to problems featuring geometrical variability and parametrized domains, namely, we propose Continuous Geometry-Aware DL-ROMs (CGA-DL-ROMs). In particular, the space-continuous nature of the proposed architecture matches the need to deal with multi-resolution datasets, which are quite common in the case of geometrically parametrized problems. Moreover, CGA-DL-ROMs are endowed with a strong inductive bias that makes them aware of geometrical parametrizations, thus enhancing both the compression capability and the overall performance of the architecture. Within this work, we justify our findings through a thorough theoretical analysis, and we practically validate our claims by means of a series of numerical tests encompassing physically-and-geometrically parametrized PDEs, ranging from the unsteady Navier-Stokes equations for fluid dynamics to advection-diffusion-reaction equations for mathematical biology.
Open AccessHandling geometrical variability in nonlinear reduced order modeling through Continuous Geometry-Aware DL-ROMs
Deep Learning-based Reduced Order Models (DL-ROMs) provide nowadays a well-established class of accurate surrogate models for complex physical systems described by parametrized PDEs, by nonlinearly compressing the solution manifold into a handful of latent coordinates. Until now, design and application of DL-ROMs mainly focused on physically parameterized problems. Within this work, we provide a novel extension of these architectures to problems featuring geometrical variability and parametrized domains, namely, we propose Continuous Geometry-Aware DL-ROMs (CGA-DL-ROMs). In particular, the space-continuous nature of the proposed architecture matches the need to deal with multi-resolution datasets, which are quite common in the case of geometrically parametrized problems. Moreover, CGA-DL-ROMs are endowed with a strong inductive bias that makes them aware of geometrical parametrizations, thus enhancing both the compression capability and the overall performance of the architecture. Within this work, we justify our findings through a thorough theoretical analysis, and we practically validate our claims by means of a series of numerical tests encompassing physically-and-geometrically parametrized PDEs, ranging from the unsteady Navier-Stokes equations for fluid dynamics to advection-diffusion-reaction equations for mathematical biology.
- Open Access
- 24 April 2024
- Published by Springer Nature in Advances in Computational Mathematics
- Vol. 50 (3) , 1-46
Error estimates for POD-DL-ROMs: a deep learning framework for reduced order modeling of nonlinear parametrized PDEs enhanced by proper orthogonal decomposition
POD-DL-ROMs have been recently proposed as an extremely versatile strategy to build accurate and reliable reduced order models (ROMs) for nonlinear parametrized partial differential equations, combining (i) a preliminary dimensionality reduction obtained through proper orthogonal decomposition (POD) for the sake of efficiency, (ii) an autoencoder architecture that further reduces the dimensionality of the POD space to a handful of latent coordinates, and (iii) a dense neural network to learn the map that describes the dynamics of the latent coordinates as a function of the input parameters and the time variable. Within this work, we aim at justifying the outstanding approximation capabilities of POD-DL-ROMs by means of a thorough error analysis, showing how the sampling required to generate training data, the dimension of the POD space, and the complexity of the underlying neural networks, impact on the solutions us to formulate practical criteria to control the relative error in the approximation of the solution field of interest, and derive general error estimates. Furthermore, we show that, from a theoretical point of view, POD-DL-ROMs outperform several deep learning-based techniques in terms of model complexity. Finally, we validate our findings by means of suitable numerical experiments, ranging from parameter-dependent operators analytically defined to several parametrized PDEs.
- 24 April 2024
- Published by Springer Nature in Advances in Computational Mathematics
- Vol. 50 (3) , 1-46
Open AccessError estimates for POD-DL-ROMs: a deep learning framework for reduced order modeling of nonlinear parametrized PDEs enhanced by proper orthogonal decomposition
POD-DL-ROMs have been recently proposed as an extremely versatile strategy to build accurate and reliable reduced order models (ROMs) for nonlinear parametrized partial differential equations, combining (i) a preliminary dimensionality reduction obtained through proper orthogonal decomposition (POD) for the sake of efficiency, (ii) an autoencoder architecture that further reduces the dimensionality of the POD space to a handful of latent coordinates, and (iii) a dense neural network to learn the map that describes the dynamics of the latent coordinates as a function of the input parameters and the time variable. Within this work, we aim at justifying the outstanding approximation capabilities of POD-DL-ROMs by means of a thorough error analysis, showing how the sampling required to generate training data, the dimension of the POD space, and the complexity of the underlying neural networks, impact on the solutions us to formulate practical criteria to control the relative error in the approximation of the solution field of interest, and derive general error estimates. Furthermore, we show that, from a theoretical point of view, POD-DL-ROMs outperform several deep learning-based techniques in terms of model complexity. Finally, we validate our findings by means of suitable numerical experiments, ranging from parameter-dependent operators analytically defined to several parametrized PDEs.
- Open Access
- 1 December 2023
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (12)
Deep learning-based surrogate models for parametrized PDEs: Handling geometric variability through graph neural networks
Mesh-based simulations play a key role when modeling complex physical systems that, in many disciplines across science and engineering, require the solution to parametrized time-dependent nonlinear partial differential equations (PDEs). In this context, full order models (FOMs), such as those relying on the finite element method, can reach high levels of accuracy, however often yielding intensive simulations to run. For this reason, surrogate models are developed to replace computationally expensive solvers with more efficient ones, which can strike favorable trade-offs between accuracy and efficiency. This work explores the potential usage of graph neural networks (GNNs) for the simulation of time-dependent PDEs in the presence of geometrical variability. In particular, we propose a systematic strategy to build surrogate models based on a data-driven time-stepping scheme where a GNN architecture is used to efficiently evolve the system. With respect to the majority of surrogate models, the proposed approach stands out for its ability of tackling problems with parameter-dependent spatial domains, while simultaneously generalizing to different geometries and mesh resolutions. We assess the effectiveness of the proposed approach through a series of numerical experiments, involving both two- and three-dimensional problems, showing that GNNs can provide a valid alternative to traditional surrogate models in terms of computational efficiency and generalization to new scenarios.
- 1 December 2023
- Published by AIP Publishing in Chaos: An Interdisciplinary Journal of Nonlinear Science
- Vol. 33 (12)
Open AccessDeep learning-based surrogate models for parametrized PDEs: Handling geometric variability through graph neural networks
Mesh-based simulations play a key role when modeling complex physical systems that, in many disciplines across science and engineering, require the solution to parametrized time-dependent nonlinear partial differential equations (PDEs). In this context, full order models (FOMs), such as those relying on the finite element method, can reach high levels of accuracy, however often yielding intensive simulations to run. For this reason, surrogate models are developed to replace computationally expensive solvers with more efficient ones, which can strike favorable trade-offs between accuracy and efficiency. This work explores the potential usage of graph neural networks (GNNs) for the simulation of time-dependent PDEs in the presence of geometrical variability. In particular, we propose a systematic strategy to build surrogate models based on a data-driven time-stepping scheme where a GNN architecture is used to efficiently evolve the system. With respect to the majority of surrogate models, the proposed approach stands out for its ability of tackling problems with parameter-dependent spatial domains, while simultaneously generalizing to different geometries and mesh resolutions. We assess the effectiveness of the proposed approach through a series of numerical experiments, involving both two- and three-dimensional problems, showing that GNNs can provide a valid alternative to traditional surrogate models in terms of computational efficiency and generalization to new scenarios.
- Open Access
Modelling the Periodic Response of Micro-Electromechanical Systems through Deep Learning-Based Approaches
We propose a deep learning-based reduced order modelling approach for micro- electromechanical systems. The method allows treating parametrised, fully coupled electromechanical problems in a non-intrusive way and provides solutions across the whole device domain almost in real time, making it suitable for design optimisation and control purposes. The proposed technique specifically addresses the steady-state response, thus strongly reducing the computational burden associated with the neural network training stage and generating deep learning models with fewer parameters than similar architectures considering generic time-dependent problems. The approach is validated on a disk resonating gyroscope exhibiting auto-parametric resonance.
Open AccessModelling the Periodic Response of Micro-Electromechanical Systems through Deep Learning-Based Approaches
We propose a deep learning-based reduced order modelling approach for micro- electromechanical systems. The method allows treating parametrised, fully coupled electromechanical problems in a non-intrusive way and provides solutions across the whole device domain almost in real time, making it suitable for design optimisation and control purposes. The proposed technique specifically addresses the steady-state response, thus strongly reducing the computational burden associated with the neural network training stage and generating deep learning models with fewer parameters than similar architectures considering generic time-dependent problems. The approach is validated on a disk resonating gyroscope exhibiting auto-parametric resonance.
- Top Cited PapersOpen Access
- 3 May 2023
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 411
Reduced order modeling of parametrized systems through autoencoders and SINDy approach: continuation of periodic solutions
- 3 May 2023
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 411
Top Cited PapersOpen AccessReduced order modeling of parametrized systems through autoencoders and SINDy approach: continuation of periodic solutions