# BDT as a Surrogate Model

- Institution: CERN
- Department: Workshops
- Taught by: Louie Dartmoor Corpe, Abdelhamid Haddad
- Year: 2025
- Language: English
- Subjects: Physics
- Source: [CERN](https://videos.cern.ch/record/3021431)
- Licence: [CERN](http://copyright.web.cern.ch)
- Who can watch: Tweens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

I will present the reinterpretation material of the CalRatio + X ATLAS analysis (arXiv:2407.09183). The analysis focuses on neutral long-lived particles decaying within the ATLAS hadronic calorimeter. The reinterpretation involves a Boosted Decision Tree (BDT) trained on truth-level variables to estimate the probability of events within the ABCD plane and assess the sensitivity of the analysis. The BDT weights, along with a Python code example, are available on HEPData to ensure reproducibility. Additionally, I will discuss how this method can be extended to other analyses, providing guidance for broader applications..

0:00:00 Slide 1

0:03:37 Slide 2

0:05:20 Slide 3

0:10:45 Slide 4

0:12:50 Slide 5

0:14:06 Slide 6

0:17:04 Slide 7

0:18:45 Slide 8

## Lectures

1. [BDT as a Surrogate Model](https://invisiblenetwork.ai/learn/cern/3021431/fdf88b985cfe), 23 min (CERN's player)