Skip to content
View saracreates's full-sized avatar

Highlights

  • Pro

Block or report saracreates

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
saracreates/README.md

💫 About me: Particle Physics x ML

Hiii! Nice to meet you! I'm Sara (she/her), a particle physicist who applies machine learning (ML) methods to address current challenges in high energy physics. I'm currently holding a research position at CERN, where I work on the next generation of particle colliders to unravel mysteries of our universe! The project is called "Future Circular Collider" (FCC) and tries to characterize the Standard Model of particle physics via probing the Higgs Boson. The Higgs Boson is an elementry particle discovered in 2012 at CERN - but the story is not over, it's a beginning: we want to understand its properties!

If you are interested in my career and background, check out my CV! (updated Jan 2025)

📌 Projects

Let me explain two of my projects:

🌠 Particle signature classification / Jet-falvor tagging at FCC (k4MLJetTagger)

At future particle colliders, one of the central physics goals is to study the Higgs boson with exceptional precision. This requires high-performance jet-flavor tagging — identifying the types of particles that jets (collections of particles flying in the same direction) originate from, especially in Higgs decays.

In this project, I have studied seven possible decay channels of the Higgs boson by training a state-of-the-art neural network, the ParticleTransformer. I have also implemented the trained model’s inference into the official software framework for future colliders, Key4HEP, making the tagger available to the whole community.

Curious about the tagger’s performance? Check out my publication.

☀️ Photon reconstruction with neural networks (PhotonRecoML)

In my bachelor thesis at the Technical University of Munich (TUM) I have performed a proof-of-principle study to test whether neural networks can help to improve the reconstruction of photons. Photons are "light particles" that can be measured with special detectors (electromagnetic calorimeters). We want to measure where exactly the photon hit the detector and how much energy it had. This is important for any physics program at high-energy experiments. I was part of the COMPASS / AMBER collaboration that runs a fixed-target experiment at CERN.

📩 How to rearch me

Feel free to email me via sara.aumiller@cern.ch or sara.aumiller@tum.de !

Popular repositories Loading

  1. PhotonRecoML PhotonRecoML Public

    Photon reconstruction with neural networks at the COMPASS/AMBER experiments at CERN

    Jupyter Notebook

  2. FullSimTagger FullSimTagger Public

    Inspection of jet-flavor tagging in full simulation at CLD.

    Python

  3. saracreates saracreates Public

    About me

  4. TaggingResults TaggingResults Public

    Results on Jet-Flavor Tagging at FCC-ee (CLD, IDEA).

    Jupyter Notebook

  5. FCC-config FCC-config Public

    Forked from HEP-FCC/FCC-config

    FCC configuration files

    Python

  6. FCCAnalyses FCCAnalyses Public

    Forked from HEP-FCC/FCCAnalyses

    Common analysis framework for the Future Circular Collider

    C++