Welcome and Introduction

• Jay Ajitbhai Sandesara (University of Wisconsin Madison (US))
• Nick Smith (Fermi National Accelerator Lab. (US))

Room: 13/2-005

Simulation-Based Inference — An Introduction

• Gilles Louppe

Room: 13/2-005

Discussion Room: 13/2-005

Scaling Neural Simulation-Based Inference at High Performance Computing for LHC analysis

• Walter Hopkins (Argonne National Laboratory (US))
• Xiangyang Ju (Lawrence Berkeley National Lab. (US))

Room: 13/2-005 This talk will present the preliminary work on scaling the Machine Learning training and Hyperparameter optimization (HPO) at High Performance Computers. We leveraged the Pytorch Lightning framework for distributed training and Ray Tune for HPO. In addition, the ML training framework automatically monitors the model’s physics performance by evaluating Neural Simulation Based Inference (NSBI) calibration and closure metrics using goodness-of-fit measures such as $\chi^2$ and Wasserstain distances. We plan to add NSBI-oriented functionality to the framework including wifi ensembling, ensembling and inference within the HPO loop. The framework currently focuses on ML training but we plan to add downstream tasks like unbinned fit.

SBI for Effective Field Theories

• Nick Smith (Fermi National Accelerator Lab. (US))

Room: 13/2-005

Closeout

• Nick Smith (Fermi National Accelerator Lab. (US))
• Jay Ajitbhai Sandesara (University of Wisconsin Madison (US))

Room: 13/2-005