Annual Meeting of 

EMS activity group on

Scientific Machine Learning

 

Dates: 24-26 March 2025

Location: Politecnico di Milano (Aula Rogers)

 

Welcome to the first edition of the Annual Meeting of EMS-AI activity group on Scientific Machine Learning which brings together scientists from mathematics, computer science, and application areas working on computational and mathematical methods in Scientific Machine Learning.

About

 

In recent years, the combination of numerical methods and machine learning has gained an ever-increasing interest as a research field within Numerical Mathematics and Scientific Computing. To further foster this development on a European level, the Topical Activity Group Scientific Machine Learning (SciML) of the European Mathematical Society (EMS TAG SciML) has been established.

This is the first edition of the Annual Meeting of EMS-AI activity group on Scientific Machine Learning which aims at bringing together scientists from mathematics, computer science, and application areas working on computational and mathematical methods in Scientific Machine Learning.

This edition will feature 18 invited talks, as well as two industrial sessions, a roundtable discussion, and a poster session. 

 

Round table discussion

The round-table will explore the interplay between machine learning, applied mathematics and scientific computing. The discussion will focus on enhancing understanding of theoretical frameworks, addressing contemporary challenges, and discussing innovative applications in the industrial and scientific sectors.

Professor Wil Schilders, President of the International Council for Industrial and Applied Mathematics (ICIAM) and Full Professor and Chair of Scientific Computing in Industry at the Department of Mathematics and Computer Science at the Technical University of Eindhoven, will chair the round table.

The panel will feature the following distinguished participants: 

• Dr. Carlo Ciancarelli, Head of Architectures & Technologies for OBS, NAV & Aerocom Dept., Thales Alenia Space S.p.A.
• Prof. Maurizio Cecconi, Director of the Residency School of Anaesthesia at Humanitas University. Vice President of the MEDTEC School. Head of the Department of Anesthesia and Intensive Care, Humanitas Research Hospital. Deputy Scientific Director for Clinical Research.
• ⁠Dr. Laura Dovera, Representative of ENI S.p.A
• ⁠Professor Luca Formaggia, President of the Italian Society of Industrial and Applied Mathematics (SIMAI)
• ⁠Dr. Felix Hildebrand, Representative of Robert Bosch GmbH
• ⁠Professor Axel Klawonn, Spokesperson for the Topical Activity Group on Scientific Machine Learning (SciML) of the European Mathematical Society (EMS)
• Professor Benedetta Morini, Representative of the Italian Mathematical Union (UMI)
• Dr. Luigi Simeone, Chief Technology Officer, Moxoff Srl
• Professor Cristina Trombetti, President of the National Institute for Advanced Mathematics (INdAM)

  

Invited Speakers

Program

Poster Session Contributions

• Belieres-Frendo Amaury, Volume-prescribed shape optimization of the Dirichlet energy with ODE flows
• Hassan Ballout, Nonlinear compressive reduced basis approximation for Parametric PDEs
• Vittorio Bauduin, Uncertainty Quantification of Turbulent Channel Flow using Physics-Informed Neural Networks
• Luca Bindini, All-in-one electrical atrial substrate indicators with deep anomaly detection
• Ivan Bioli, Accelerating PINNs Training with Efficient Randomized Preconditioners
• Giovanni Bocchi, Explainable Machine Learning with Group Equivariant Non-Expansive Operators (GENEOs). An industrial application to Protein Pocket Detection.
• Giacomo Bottacini, Approximate Bayesian Computing: Simulation-Based Inference via Normalizing Flows and Variational Autoencoders
• Benedikt Brantner, Structure-Preserving Neural Networks for Reduced Order Modeling
• Simone Brivio, Mitigating the adverse effects of data scarcity through pre-trained physics-informed DL-ROMs
• Matteo Caldana, Discovering Artificial Viscosity Models for Discontinuous Galerkin Approximation of Conservation Laws using Physics-Informed Machine Learning
• Filippo Camellini, Majorization-Minimization for multiclass classification in a big data scenario
• Davide Carrara, Implicit Neural Field Reconstruction on Complex Shapes from Scattered Data
• Paolo Conti, Multi-fidelity reduced-order surrogate modeling
• Mattia Corti, Polytopal mesh agglomeration strategies and applications to brain pathology
• Davide Elia De Falco, ELM training for Physics-Informed Neural Networks
• Alessandra De Rossi, Community detection algorithms for signal approximation on graphs
• Marco Dell'Orto, Variationally Mimetic Operator Networks
• Nunzio Dimola, A Neural Preconditioner for Mixed-Dimensional PDEs
• Felix Döppel, Physically Consistent Neural Network Surrogates of Chemical Reactors
• Mukul Dwivedi, Stability and convergence of fractional physics informed neural network based numerical method for the fractional Korteweg-de Vries equation
• Nicola Farenga, On latent dynamics learning in nonlinear reduced order modeling
• Carlotta Filippin, Nonlinear reduced-order modeling with a Graph Convolutional Autoencoder for time-domain electromagnetics
• Nicola Rares Franco, Deep orthogonal decomposition: an adaptive basis approach to dimensionality reduction
• Massimiliano Ghiotto, HyperNOs: Automated and Parallel Hyperparameter Optimization Library for Neural Operators​​​​​​​
• Viviana Giorgi, On the Performance of Data-driven Dynamic Models for Temperature Compensation on Bridge Monitoring Data
• Ion Victor Gosea, Enabling digital twins in process and chemical engineering through scientific machine learning
• Sebastian Götschel, ML-enhanced numerics for science and engineering
• Adeeba Haider, Optimizing the shape parameters in applications of RBF-PU method
• Alexander Heinlein, High-resolution image segmentation with U-Net-based segmentation CNN on multiple GPUs
• Matthias Höfler, Estimating active stress in cardiac biomechanical models based on physics-informed neural networks
• Pinki Khatun, A Parameterized enhanced Shift-Splitting Preconditioner for Efficiently Solving Three-by-Three Block Saddle Point Problems
• Jona Klemenc, Trade-Off Invariance Principle for Minimizers of Regularized Functionals
• Prashant Kumar, Prediction of Shocks using a Consistent PINN-based Approach
• Federico Lanteri, Learning the dynamics of free surface fluids problems through a mesh-based Graph Neural Network surrogate model
• Frédérique Lecourtier, Enriching continuous Lagrange finite element approximation spaces using neural networks
• Fanny Lehmann, Learning from the latent space of weather foundation models
• Lucas Mager, Machine learning-based surrogate models for an efficient homogenization of open-porous materials
• Caterina Millevoi, Time-Dependent Parameter Estimation and Short-Term Forecasting in Compartmental Epidemiological Models Using a Reduced-Split PINN Approach
• Johannes Müller, Kronecker-Factored Approximate Curvature for Physics-Informed Neural Networks
• Luca Pellegrini, Learning Ionic Models Dynamics Using Fourier Neural Operators
• Julia Pelzer, Scalable Surrogates for Groundwater Flow Simulations
• Manvendra Pratap Rajvanshi, ML for Circular Hydraulic Jump: A Case Study on ML-assisted Simulation Techniques in Tackling Shock and Turbulence in Shallow Water Equations
• Swarup Kumar Sahoo, Scientific Machine Learning-Driven Analysis of Shallow Wave Dynamics with Dispersive Effects for Geophysical Ocean Modeling
• Simon Schneider, Estimatable variation neural networks and their application to scalar hyperbolic conservation laws
• Claire Schnoebelen, Geometric structure preserving learning for Hamiltonian PDE and ODE
• Janne Siipola, Better Sampling with Gradient
• Luca Sosta, An Interpretable Graph-Based Digital Twin towards Satellite Thermal Control and Predictive Maintenance
• Christopher Straub, Hard-constraining techniques and architectures in physics-informed neural networks for silicidation simulations
• Khoi Mai Tieu, Large Language Models for Market Sentiment Analysis and Investigating Their Impact on Stock Movements
• Matteo Tomasetto, Reduced order modeling with shallow recurrent decoder networks
• Matteo Torzoni, Predictive digital twins for structural health monitoring
• Ilaria Trombini, Variable metric proximal stochastic gradient methods with additional sampling
• Piermario Vitullo, Domain decomposition with nonlinear model order reduction for multiscale mixed-dimensional problems
• Vincenzo Vocca, A Physics-Informed Neural Network Approach for Predicting Soil Microbiota Growth
• Vincenzo Vocca, Enhancing Liquidity Provision in Automated Market Makers Using Neural Networks
• Janine Weber, Divide, learn, and conquer in image classification
• Filippo Zacchei, MFDA: a multifidelity approach for efficient Bayesian inverse problems based on fusion and filtering
• Linying Zhang, Shape-informed surrogate models based on signed distance function domain encoding
• Giovanni Ziarelli, Learning Activation and Repolarization Times with Operator Learning
•  
•  
•  
•  

Organizing Committee

P.F. Antonietti, S. Pagani, F. Regazzoni, M. Verani (chair), P. Zunino

MOX - Laboratory for Modeling and Scientific Computing, Department of Mathematics, Politecnico di Milano, Milan, Italy

 

 

Scientific Committee

• Peter Benner, Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany
• Bruno Després, Jacques-Louis Lions Laboratory, Sorbonne University, France
• Victorita Dolean, Dept. of Mathematics and Computer Science, Eindhoven University of Technology, Eindhoven, The Netherlands
• Virginie Ehrlacher (Co-Spokesperson), Ecole des Ponts Paris Tech, Paris, France
• Alexander Heinlein, Delft University of Technology, Delft, The Netherlands
• Jan Hesthaven, Chair of Computational Mathematics and Simulation Science, EPFL, Lausanne, Switzerland
• Axel Klawonn (Spokesperson), Chair of Applied Mathematics and Scientific Computing, University of Cologne, Cologne, Germany
• Rolf Krause, Chair for Advanced Scientific Computing, Euler Institute, Università della Svizzera italiana, Lugano, Switzerland
• Olga Mula (Co-Spokesperson), Center for Analysis, Scientific Computing and Applications, Eindhoven University of Technology, Eindhoven, The Netherlands
• Ozan Öktem, Numerik, Optimering, and System, KTH Stockholm, Stockholm, Sweden
• Gianluigi Rozza, SISSA, Trieste, Italy
• Benjamin Sanderse, Department of Scientific Computing, CWI, Amsterdam, The Netherlands
• Marco Verani, MOX - Laboratory for Modeling and Scientific Computing, Department of Mathematics, Politecnico di Milano, Milan, Italy
• Andrea Walther, Department of Mathematics, Humboldt-University of Berlin, Berlin, Germany

 

 

 Venue

Aula Rogers

Building 11, Via Ampère, 2 - 20133 - Milano (MI) 

The workshop is located near Piola subway station on Metro Line 2 (MM2 Green Line). Upon arrival at Piola station, please proceed to the left exit. Continue on via Francesco d'Ovidio for ~30 metres, then turn left onto the walkway towards Via Ampère. Once arrived at Via Ampère, make a right turn, which will lead you to the Faculty of Architecture entrance on your left. 

For those arriving by train, the closest station is Milano Lambrate, which is a 15-minute walk away. The average travel time from Milan Central Station is 16 minutes by metro (MM2 Green Line), while the journey from Milan Cadorna takes 22 minutes. The Milano Metro map is available at this link.

 

Poster session Venue: Aula Vetrata - Trifoglio

Building 13, Via Bonardi, 9 - 20133 - Milano (MI) 

 

Accomodation

Some hotels conveniently located near the conference venue are: 

21 House of Stories - Milano Città Studi 

Hotel Lombardia

Hotel Gamma

Hotel San Francisco

Sponsor and Acknowledgments

 

 

This event is supported by: