Nico Hoffmann

Nico Hoffmann

Group leader

Publications

Continual learning autoencoder training for a particle-in-cell simulation via streaming

Published in Machine Learning and the Physical Sciences @ NeurIPS 2022, 2022

Continual learning autoencoder training for a particle-in-cell simulation via streaming

Recommended citation: Stiller P., Makdani V., Pöschel F. , Richard P., Debus A., Bussmann M., Hoffmann N. (2022). Continual learning autoencoder training for a particle-in-cell simulation via streaming. Machine Learning and the Physical Sciences workshop @ NeurIPS 2022. https://arxiv.org/abs/2211.04770

Learning Electron Bunch Distribution along a FEL Beamline by Normalising Flows

Published in Machine Learning and the Physical Sciences @ NeurIPS 2022, 2022

Learning Electron Bunch Distribution along a FEL Beamline by Normalising Flows

Recommended citation: Willmann A.,Cabadag J.C.,Chang Y.-Y.,Pausch R.,Ghaith A.,Debus A.,Irman A., Bussmann M., Schramm U., Hoffmann N. (2022). Learning Electron Bunch Distribution along a FEL Beamline by Normalising Flows. Machine Learning and the Physical Sciences @ NeurIPS 2022. https://ml4physicalsciences.github.io/2022/files/NeurIPS_ML4PS_2022_98.pdf

Invertible Surrogate Models: Joint Surrogate Modelling and Reconstruction of Laser Wakefield Acceleration by Invertible Neural Networks

Published in Simulation with Deep Learning @ ICLR, 2021

We will be introducing invertible surrogate models that approximate complex forward simulation of the physics involved in laser plasma accelerators: iLWFA. The bijective design of the surrogate model also provides all means for reconstruction of experimentally acquired diagnostics. The quality of our invertible laser wakefield acceleration network will be verified on a large set of numerical LWFA simulations.

Recommended citation: Bethke, F., Pausch, R., Stiller, P., Debus, A., Bussmann, M., Hoffmann, N. (2021). Invertible Surrogate Models: Joint Surrogate Modelling and Reconstruction of Laser Wakefield Acceleration by Invertible Neural Networks. Simulation with Deep Learning @ ICLR. https://arxiv.org/abs/2106.00432

Data-Driven Shadowgraph Simulation of a 3D Object

Published in Simulation with Deep Learning @ ICLR, 2021

We propose a deep neural network based surrogate model for a plasma shadowgraph - a technique for visualization of perturbations in a transparent medium. We are substituting the numerical code by a computationally cheaper projection based surrogate model that is able to approximate the electric fields at a given time without computing all preceding electric fields as required by numerical methods. This means that the projection based surrogate model allows to recover the solution of the governing 3D partial differential equation, 3D wave equation, at any point of a given compute domain and configuration without the need to run a full simulation. This model has shown a good quality of reconstruction in a problem of interpolation of data within a narrow range of simulation parameters and can be used for input data of large size.

Recommended citation: Willmann, A., Stiller, P., Debus, A., Irman, A., Pausch, R., Chang, Y.-Y.,Bussmann, M., Hoffmann, N. (2021). Data-Driven Shadowgraph Simulation of a 3D Object. Simulation with Deep Learning @ ICLR. https://arxiv.org/abs/2106.00317

NAM: Normalization-based Attention Module

Published in ImageNet PPF @ NeurIPS 2021, 2021

we propose a novel normalization-based attention module (NAM), which suppresses less salient weights. It applies a weight sparsity penalty to the attention modules, thus, making them more computational efficient while retaining similar performance.

Recommended citation: Liu, Y., Shao, Z., Teng, Y., Hoffmann N. (2021). NAM: Normalization-based Attention Module. ImageNet PPF @ NeurIPS 2021. https://arxiv.org/abs/2111.12419