Intro

• Lukas Heinrich (TUM)

Room: X5/1.1.18b (big seminar room)

Neural Simulation-Based Inference for the KeV-Scale Sterile Neutrino Search with TRISTAN at KATRIN (15'+5')

• Luca Fallböhmer

Room: X5/1.1.18b (big seminar room) **Abstract:** Following the completion of its neutrino mass measurement program at the end of 2025, the KATRIN experiment aims to probe keV-scale sterile neutrinos by analyzing the full tritium beta decay spectrum with a novel detector system, TRISTAN. Leveraging KATRIN's high source activity, this search is sensitive to mixing amplitudes at the parts-per-million level. However, extracting a potential sterile neutrino signature is challenging, as it relies on detailed modeling of the observed tritium spectrum and requires computationally intensive Monte Carlo simulations. To address this challenge, we implement neural simulation-based inference using normalizing flows to approximate the underlying probability density of the physics simulation. We demonstrate that continuous normalizing flows trained via conditional flow-matching enable modeling of experimental spectra and discuss the challenges involved.

Vertexing with ML for discovering ALPS (15'+5')

• Daniel Grewe

Room: X5/1.1.18b (big seminar room)

Characterizing the emergence of the complex molecular, dynamical molecular gas web in galaxies (15'+5')

• Andreas Burkert (LMU)

Room: X5/1.1.18b (big seminar room)

When MAGIC met IceCube: Gamma-Ray Astronomy with Neutrino Event Reconstruction Techniques (15'+5')

• Jarred Green (MPP)

Room: X5/1.1.18b (big seminar room)

Differentiable quantum Monte Carlo simulations for inference in Heavy Ion Collisions (15'+5')

• Tom Magorsch (TU Munich)

Room: X5/1.1.18b (big seminar room)

Likelihood-Free Conditional Inference for LEGEND-200 LAr Anti-Coincidence Cut (15'+5')

• Niko Nanda Putra Nila Lay (Technical University of Munich)

Room: X5/1.1.18b (big seminar room)

Simulation-Based Inference for eROSITA Galaxy Cluster Cosmology (15'+5')

• Silas Zelmer (MPE)

Room: X5/1.1.18b (big seminar room) **Abstract:** We present a simulation-based inference (SBI) framework for cosmological parameter estimation from eROSITA galaxy cluster catalogs. Our approach combines a realistic forward-modeling pipeline with a set-based neural network and normalizing flows to leverage the full cluster-level information content. Using mock catalogs matching the size and selection properties of the eRASS1 sample, we demonstrate accurate and well-calibrated recovery of cosmological parameters in the presence of survey systematics. Despite the moderate sample size (~3,300 clusters), our method achieves constraints on Ωₘ and σ₈ comparable to traditional analyses based on significantly larger datasets. This highlights the potential of SBI methods for next-generation large-scale structure studies and the analysis of high-dimensional survey data. https://arxiv.org/abs/2606.27439

Simulation-based inference of galaxy clustering and population (15'+5')

• Silvan Fischbacher

Room: X5/1.1.18b (big seminar room)

Gradient estimators for parameter inference in discrete stochastic kinetic models (15'+5')

• Ludwig Burger

Room: X5/1.1.18b (big seminar room)

Parameter-Free Light Antinuclei Coalescence: Bridging Colliders and Cosmic Rays with Neural-Flow Surrogates (15'+5')

• Maximilian Korwieser (TUM)

Room: X5/1.1.18b (big seminar room)

Why do we get old (15'+5')

• Onurcan Bektas

Room: X5/1.1.18b (big seminar room)

Closeout / discussion Room: X5/1.1.18b (big seminar room)