Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Incorporating Physical Priors into Weakly Supervised Anomaly Detection

Journal Article · · Physical Review Letters
DOI:https://doi.org/10.1103/8259-wt5p· OSTI ID:2583310
 [1];  [2];  [3]
  1. Univ. of Wisconsin, Madison, WI (United States); Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
  2. Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
  3. Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States); Stanford Univ., CA (United States); SLAC National Accelerator Laboratory (SLAC), Menlo Park, CA (United States)

We propose a new machine-learning-based anomaly detection strategy for comparing data with a background-only reference (a form of weak supervision). The sensitivity of previous strategies degrades significantly when the signal is too rare or there are many unhelpful features. Our prior-assisted weak supervision (PAWS) method incorporates information from a class of signal models to significantly enhance the search sensitivity of weakly supervised approaches. As long as the true signal is in the prespecified class, PAWS matches the sensitivity of a dedicated, fully supervised method without specifying the exact parameters ahead of time. On the benchmark LHC Olympics anomaly detection dataset, our mix of semisupervised and weakly supervised learning is able to extend the sensitivity over previous methods by a factor of 10 in cross section. Furthermore, if we add irrelevant (noise) dimensions to the inputs, classical methods degrade by another factor of 10 in cross section while PAWS remains insensitive to noise. This new approach could be applied in a number of scenarios and pushes the frontier of sensitivity between completely model-agnostic approaches and fully model-specific searches.

Research Organization:
Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
Sponsoring Organization:
USDOE Office of Science (SC), Basic Energy Sciences (BES). Scientific User Facilities (SUF); USDOE Office of Science (SC), Office of Workforce Development for Teachers & Scientists (WDTS)
Grant/Contract Number:
AC02-05CH11231; SC0017647
OSTI ID:
2583310
Journal Information:
Physical Review Letters, Journal Name: Physical Review Letters Journal Issue: 2 Vol. 135; ISSN 1079-7114; ISSN 0031-9007
Publisher:
American Physical Society (APS)Copyright Statement
Country of Publication:
United States
Language:
English

References (44)

Maximizing boosted top identification by minimizing N-subjettiness journal February 2012
DELPHES 3: a modular framework for fast simulation of a generic collider experiment journal February 2014
Improving the performance of weak supervision searches using transfer and meta-learning journal February 2024
Anomaly detection in the presence of irrelevant features journal February 2024
Identifying boosted objects with N-subjettiness journal March 2011
Quasi anomalous knowledge: searching for new physics with embedded knowledge journal June 2021
Classification without labels: learning from mixed samples in high energy physics journal October 2017
An introduction to PYTHIA 8.2 journal June 2015
Dispelling the N3 myth for the kt jet-finder journal September 2006
Machine learning in the search for new fundamental physics journal May 2022
The frontier of simulation-based inference journal May 2020
PYTHIA 6.4 physics and manual journal May 2006
The anti- k t jet clustering algorithm journal April 2008
The LHC Olympics 2020 a community challenge for anomaly detection in high energy physics journal December 2021
New features in Delphes 3 journal May 2015
Masked particle modeling on sets: towards self-supervised high energy physics foundation models journal September 2024
via machinae: Searching for stellar streams using unsupervised machine learning journal November 2021
Weakly supervised anomaly detection in the Milky Way journal November 2023
skycurtains: model-agnostic search for stellar streams with Gaia data journal November 2024
Via Machinae 2.0: Full-Sky, Model-Agnostic Search for Stellar Streams in Gaia DR2 journal February 2024
Anomaly detection with density estimation journal April 2020
Simulation assisted likelihood-free anomaly detection journal May 2020
Simulation-assisted decorrelation for resonant anomaly detection journal August 2021
Classifying anomalies through outer density estimation journal September 2022
Flow-enhanced transportation for anomaly detection journal May 2023
Resonant anomaly detection without background sculpting journal June 2023
Anomaly detection search for new resonances decaying into a Higgs boson and a generic new particle X in hadronic final states using s=13  TeV pp collisions with the ATLAS detector journal September 2023
Tree-based algorithms for weakly supervised anomaly detection journal February 2024
Full phase space resonant anomaly detection journal March 2024
Method to simultaneously facilitate all jet physics tasks journal March 2025
Solving key challenges in collider physics with foundation models journal March 2025
Extending the search for new resonances with machine learning journal January 2019
Learning new physics from a machine journal January 2019
Anomaly Detection for Resonant New Physics with Machine Learning journal December 2018
Dijet Resonance Search with Weak Supervision Using s = 13     TeV p p Collisions in the ATLAS Detector journal September 2020
Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at s=13  TeV with the ATLAS Detector journal February 2024
FastJet user manual: (for version 3.0.2) journal March 2012
Parameterized neural networks for high-energy physics journal April 2016
Learning multivariate new physics journal January 2021
Learning new physics from an imperfect machine journal March 2022
The interplay of machine learning-based resonant anomaly detection methods journal March 2024
The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collider journal January 2022
Anomaly Awareness journal August 2023
CURTAINs flows for flows: Constructing unobserved regions with maximum likelihood estimation journal August 2024

Similar Records

The LHC Olympics 2020 a community challenge for anomaly detection in high energy physics
Journal Article · Mon Dec 06 23:00:00 EST 2021 · Reports on Progress in Physics · OSTI ID:1863790