IRN Terascale @ Heidelberg

Session

Methods and Tools

20 oct. 2026, 14:00

International Academic Forum - Heidelberg University

27. Anomaly detection - Review

Louis Moureaux (Universität Hamburg)

20/10/2026 14:00

Methods and Tools

Despite strong motivation from theory and an intensive search
programme, no particle beyond the standard model has been
discovered at the LHC. Since model specific searches cannot
test every possible signal model, a broader approach capable
of testing many hypotheses at once is needed. Anomaly
detection refers to the set of machine-learning methods
specifically designed for...

16. Autoencoder-based anomaly detection and NAE

Lorenz Vogel (ITP, Heidelberg University)

20/10/2026 14:30

Methods and Tools

Unsupervised anomaly detection with autoencoders is a promising data-driven and model-agnostic approach for new physics searches at the LHC. However, although neural network-based anomaly scores have transformative potential, they suffer from a lack of statistical interpretability. The normalized autoencoder (NAE) provides a probabilistic interpretation of the standard bottleneck architecture,...

15. The Information Geometry of Learned Latent Representations

Rebecca Maria Kuntz (Astronomisches Rechen-Institut, Universität Heidelberg)

20/10/2026 14:50

Methods and Tools

Many classification and reconstruction tasks in physics rely on learned latent representations of the data. When networks are trained with a notion of locality, they encode task-specific similarity as closeness in the latent space. Differential geometry, particularly information geometry, is a powerful tool to uncover the learned information in these latent representations and thereby retrace...

14. Neural Control Variates at LO and NLO

Rebecca Revelli (Institute for theoretical physics Heidelberg)

20/10/2026 15:45

Methods and Tools

We employ neural control variates to minimize the range of event weights and avoid negative weights for phase-space integration and event generation. A signed control variate, built from two normalizing flows, fulfills both tasks. Combined with neural importance sampling, it significantly reduces the computational cost of LO and NLO predictions. For the NLO case, our conditional neural control...

18. Sequential Simulation-Based Inference for the LHC

Ranit Das (Universität Heidelberg)

20/10/2026 16:05

Methods and Tools

Simulation-based inference (SBI) is a powerful tool for likelihood-free parameter estimation, but often requires large numbers of simulated events. At the LHC, where event generation can be computationally expensive, particularly with restrictive generator-level cuts and detector simulation, this can become a major bottleneck. We demonstrate sequential SBI for the inference of resonance masses...

20. MadAgents and Sfitter analyses

Daniel Schiller (Institute for Theoretical Physics, Heidelberg University)

20/10/2026 16:25

Methods and Tools

We present an effective and communicative set of agents working with MadGraph and SFitter. Agentic installation, learning-by-doing training, and user support provide easy access to state-of-the-art simulations and accelerate LHC research. We show how MadAgents interact with inexperienced and advanced users, automatize event generation and autonomous simulation campaigns, and update a global...