mlsanityData to 7 Oct 2026

TRANSIT your events into a new mass: Fast background interpolation for weakly-supervised anomaly searches

arXiv 2503.04342 ↗First posted 6 Mar 2025hep-ph cs.LG hep-excited by 0

Not on the map. Why?

Abstract (arXiv)

We introduce a new model for conditional and continuous data morphing called TRansport Adversarial Network for Smooth InTerpolation (TRANSIT). We apply it to create a background data template for weakly-supervised searches at the LHC. The method smoothly transforms sideband events to match signal region mass distributions. We demonstrate the performance of TRANSIT using the LHC Olympics R&D dataset. The model captures non-linear mass correlations of features and produces a template that offers a competitive anomaly sensitivity compared to state-of-the-art transport-based template generators. Moreover, the computational training time required for TRANSIT is an order of magnitude lower than that of competing deep learning methods. This makes it ideal for analyses that iterate over many signal regions and signal models. Unlike generative models, which must learn a full probability density distribution, i.e., the correlations between all the variables, the proposed transport model only has to learn a smooth conditional shift of the distribution. This allows for a simpler, more efficient residual architecture, enabling mass uncorrelated features to pass the network unchanged while the mass correlated features are adjusted accordingly. Furthermore, we show that the latent space of the model provides a set of mass decorrelated features useful for anomaly detection without background sculpting.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 130 references that are in the map. First posted on arXiv 6 Mar 2025.

Builds on

32 of 130 references.

References · 32

  1. 2020Anomaly Detection with Density Estimationcited by 71
  2. 2020Anomaly Detection With Conditional Variational Autoencoderscited by 46
  3. 2017Classification without labels: Learning from mixed samples in high energy physicscited by 42
  4. 2018Variational Autoencoders for New Physics Mining at the Large Hadron Collidercited by 40
  5. 2020Variational Autoencoders for Anomalous Jet Taggingcited by 33
  6. 2021Autoencoders for unsupervised anomaly detection in high energy physicscited by 32
  7. 2021The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collidercited by 28
  8. 2020Simulation Assisted Likelihood-free Anomaly Detectioncited by 27
  9. 2021Topological Obstructions to Autoencodingcited by 20
  10. 2021Better Latent Spaces for Better Autoencoderscited by 20
  11. 2022Learning new physics efficiently with nonparametric methodscited by 20
  12. 2021Bump Hunting in Latent Spacecited by 19
20 more
  1. 2020Adversarially Learned Anomaly Detection on CMS Open Data: re-discovering the top quarkcited by 18
  2. 2021Challenges for Unsupervised Anomaly Detection in Particle Physicscited by 16
  3. 2021Online-compatible Unsupervised Non-resonant Anomaly Detectioncited by 15
  4. 2020Simulation-Assisted Decorrelation for Resonant Anomaly Detectioncited by 13
  5. 2021Comparing Weak- and Unsupervised Methods for Resonant Anomaly Detectioncited by 13
  6. 2020Unsupervised in-distribution anomaly detection of new physics through conditional density estimationcited by 12
  7. 2023High-dimensional and Permutation Invariant Anomaly Detectioncited by 12
  8. 2020Anomaly Awarenesscited by 10
  9. 2023CURTAINs Flows For Flows: Constructing Unobserved Regions with Maximum Likelihood Estimationcited by 10
  10. 2023Improving new physics searches with diffusion models for event observables and jet constituentscited by 10
  11. 2021Autoencoders for Semivisible Jet Detectioncited by 9
  12. 2023Unravelling physics beyond the standard model with classical and quantum anomaly detectioncited by 9
  13. 2022"Flux+Mutability": A Conditional Generative Approach to One-Class Classification and Anomaly Detectioncited by 7
  14. 2022Neural Embedding: Learning the Embedding of the Manifold of Physics Datacited by 7
  15. 2022Null Hypothesis Test for Anomaly Detectioncited by 7
  16. 2022Quantum-probabilistic Hamiltonian learning for generative modelling & anomaly detectioncited by 7
  17. 2021New Methods and Datasets for Group Anomaly Detection From Fundamental Physicscited by 6
  18. 2023Fast Particle-based Anomaly Detection Algorithm with Variational Autoencodercited by 6
  19. 2023TURBO: The Swiss Knife of Auto-Encoderscited by 3
  20. 2024SIGMA: Single Interpolated Generative Model for Anomaliescited by 2

Built on by

Not cited yet.