Seminar by Prof. Sibo Cheng at the Faculty of Mathematics and Physics

University of Ljubljana, Faculty of Mathematics and Physics, Jadranska ulica 19, Ljubljana

Prof. Cheng is a Junior Professor at École Nationale des Ponts et Chaussées (ENPC), Institut Polytechnique de Paris in France. He will talk about Hybrid Data Assimilation and Machine Learning algorithms for Sparse Observations in Geoscience and Air Pollution Modeling .

Abstract:

Recently, significant efforts have been made to merge Data Assimilation (DA) and Deep Learning (DL) approaches, with objectives including parameter calibration, reduced-order surrogate modeling, and model error correction. We present our recent work on multivariate field reconstruction using sparse observation data, enabled by the development of our new Python package, TorchDA, which seamlessly integrates DL predictive functions into DA workflows. This new package enables the implementation of the Kalman Filter, the Ensemble Kalman Filter, 3D-Variational, and 4D-Variational algorithms in PyTorch, allowing flexible algorithm selection based on application requirements. The main advantages of TorchDA, compared to existing Python-based DA packages, are twofold: (i) it can handle trained neural networks as forward and transformation functions, eliminating the need for explicit transformation and forward functions, and (ii) it allows GPU acceleration for online optimization, particularly in variational DA methods. Numerical results will be demonstrated using two-dimensional CFD models, the open-access WeatherBench dataset, and a hydrology application.

  1. References:
    *Cheng, S., et al., 2023. Machine learning with data assimilation and uncertainty quantification for dynamical systems: a review. IEEE/CAA Journal of Automatica Sinica, 10(6), pp.1361-1387.
    *Cheng, S., Min, J., Liu, C. and Arcucci, R., 2025. TorchDA: A Python package for performing data assimilation with deep learning forward and transformation functions. Computer Physics Communications, 306, p.109359.

Short bio: Sibo Cheng is currently a Junior Professor (Chaire de Professeur Junior) at CEREA, École Nationale des Ponts et Chaussées (ENPC), Institut Polytechnique de Paris in France. His work focuses on machine learning for dynamical systems, reduced-order surrogate models (digital twins), and inverse modeling (parameter calibration and data assimilation) for environmental science and physics, with a wide range of applications including geosciences (wildfire & air pollution) and fluid dynamics. With an engineering degree and an MSc in applied mathematics, he completed his Ph.D. at LISN, University Paris-Saclay, France, in 2020. From 2020 to 2024, he was a research associate at the Data Science Institute of Imperial College London.

After the seminar, we will be hosting a small gathering with light refreshments. It will be a wonderful opportunity to have discussions and to connect with fellow attendees.