Vinicius

Vinicius

Postdoctoral scholar at the Lawrence Berkeley National Lab. developing machine learning methods for natural sciences.

Unsupervised clustering for collider physics

Published in Phys.Rev.D 103 (2021) 9, 092007, 2021

Recommended citation: Vinicius Mikuni and Florencia Canelli Phys. Rev. D *103*, 092007 https://journals.aps.org/prd/abstract/10.1103/PhysRevD.103.092007

We propose a new method for unsupervised clustering for collider physics named UCluster, where information in the embedding space created by a neural network is used to categorize collision events into different clusters that share similar properties. We show how this method can be developed into an unsupervised multiclass classification of different processes and applied in the anomaly detection of events to search for new physics phenomena at colliders.

Download paper here

Algorithm implementation

Recommended citation:

@article{PhysRevD.103.092007,
  title = {Unsupervised clustering for collider physics},
  author = {Mikuni, Vinicius and Canelli, Florencia},
  journal = {Phys. Rev. D},
  volume = {103},
  issue = {9},
  pages = {092007},
  numpages = {12},
  year = {2021},
  month = {May},
  publisher = {American Physical Society},
  doi = {10.1103/PhysRevD.103.092007},
  url = {https://link.aps.org/doi/10.1103/PhysRevD.103.092007}
}