Research interests

  • Simulation-based (likelihood-free) inference: machine learning in the context of (scientific) simulator models
  • Probabilistic models, in particular normalizing flows
  • The embedding of domain knowledge (structure, invariances, equivariances) into ML models and algorithms
  • (Bayesian) statistics
  • Machine learning for problems in particle physics and astrophysics

Papers

For a more or less complete list, see Google Scholar, ADS, INSPIRE, or Papers with Code. Here I want to highlight a few favorites:

Probabilistic / generative models

  • Johann Brehmer and Kyle Cranmer:
    Flows for simultaneous manifold learning and density estimation.
    NeurIPS 2020. [arXiv] [conference] [bibtex] [code]
    (Also: Spotlight talk at the ICML workshop on Invertible Neural Networks, Normalizing Flows, and Explicit Likelihood Models (2020). [workshop])

Simulation-based inference methods

  • Kyle Cranmer, Johann Brehmer, and Gilles Louppe:
    The frontier of simulation-based inference.
    Proceedings of the National Academy of Science 117 (2020). [arXiv] [journal] [bibtex]
  • Johann Brehmer, Gilles Louppe, Juan Pavez, and Kyle Cranmer:
    Mining gold from implicit models to improve likelihood-free inference.
    Proceedings of the National Academy of Science 117 (2020). [arXiv] [journal] [bibtex]
    (Also: NeurIPS workshop on Machine Learning and the Physical Sciences (2019). [workshop])
  • Markus Stoye, Johann Brehmer, Gilles Louppe, Juan Pavez, and Kyle Cranmer:
    Likelihood-free inference with an improved cross-entropy estimator.
    NeurIPS workshop on Machine Learning and the Physical Sciences (2019). [arXiv] [workshop] [bibtex]

Simulation-based inference in physics

  • Johann Brehmer and Kyle Cranmer:
    Simulation-based inference methods for particle physics.
    To appear in Artificial Intelligence for Particle Physics. [arXiv] [bibtex]
  • Johann Brehmer, Siddharth Mishra-Sharma, Joeri Hermans, Gilles Louppe, and Kyle Cranmer:
    Mining for Dark Matter Substructure: Inferring subhalo population properties from strong lenses with machine learning.
    The Astrophysical Journal 886 (2019). [arXiv] [journal] [bibtex] [code]
  • Johann Brehmer, Felix Kling, Irina Espejo, and Kyle Cranmer:
    MadMiner: Machine learning-based inference for particle physics.
    Computing and Software for Big Science 4 (2020). [arXiv] [journal] [bibtex] [code]
  • Johann Brehmer, Kyle Cranmer, Gilles Louppe, and Juan Pavez:
    Constraining effective field theories with machine learning.
    Physical Review Letters 121 (2018). [arXiv] [journal] [bibtex] [code]
  • Johann Brehmer, Kyle Cranmer, Gilles Louppe, and Juan Pavez:
    A guide to constraining effective field theories with machine learning.
    Physical Review D 98 (2018). [arXiv] [journal] [bibtex] [code]

More deep learning for physics

  • Johann Brehmer, Sebastian Macaluso, Duccio Pappadopulo, and Kyle Cranmer:
    Hierarchical clustering in particle physics through reinforcement learning.
    NeurIPS workshop on Machine Learning and the Physical Sciences (2020). [arXiv] [bibtex] [code]
  • Isaac Henrion, Johann Brehmer, Joan Bruna, Kyunghun Cho, Kyle Cranmer, Gilles Louppe, and Gaspar Rochette:
    Neural Message Passing for Jet Physics.
    NeurIPS workshop on Deep Learning for the Physical Sciences (2017). [workshop]

Information geometry, experimental design, and particle physics walk into a bar

  • Johann Brehmer, Kyle Cranmer, Felix Kling, Tilman Plehn:
    Better Higgs Measurements Through Information Geometry.
    Physical Review D 95 (2017). [arXiv] [journal] [bibtex]

Particle physics theory

  • Johann Brehmer, Ayres Freitas, David Lopez-Val, Tilman Plehn:
    Pushing Higgs Effective Theory to its Limits.
    Physical Review D 93 (2016). [arXiv] [journal] [bibtex]

Code

Together with Felix Kling, Irina Espejo, Sinclert Perez, and Kyle Cranmer, I have developed the MadMiner library, which automates machine learning–powered simulation-based inference techniques for particle physics experiments.

My GitHub page contains the code used for most of my papers. I also collect tools and tips that I find useful.

Talks

PDF slides from my talks can also be found on GitHub.

Contact

Write me at mail@johannbrehmer.de.

Not work

I like travelling and taking pictures.