Creating Simple, Interpretable Anomaly Detectors for New Physics in Jet Substructure

Abstract

Anomaly detection with convolutional autoencoders is a popular method to search for new physics in a model-agnostic manner. These techniques are powerful, but they are still a "black box," since we do not know what high-level physical observables determine how anomalous an event is. To address this, we adapt a recently proposed technique by Faucett et al., which maps out the physical observables learned by a neural network classifier, to the case of anomaly detection. We propose two different strategies that use a small number of high-level observables to mimic the decisions made by the autoencoder on background events, one designed to directly learn the output of the autoencoder, and the other designed to learn the difference between the autoencoder's outputs on a pair of events. Despite the underlying differences in their approach, we find that both strategies have similar ordering performance as the autoencoder and independently use the same six high-level observables. From there, we compare the performance of these networks as anomaly detectors. We find that both strategies perform similarly to the autoencoder across a variety of signals, giving a nontrivial demonstration that learning to order background events transfers to ordering a variety of signal events.

17 pages, 8 figures; Corrections made for p_T, m calculations; analysis rerun with conclusions largely unchanged. Published version

References in corpus (13)

• The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
• An Introduction to PYTHIA 8.2
• Power Counting to Better Jet Observables
• Gaining (Mutual) Information about Quark/Gluon Discrimination
• New Angles on Energy Correlation Functions
• A generic anti-QCD jet tagger
• Jet Substructure Without Trees
• Autoencoders on FPGAs for real-time, unsupervised new physics detection at 40 MHz at the Large Hadron Collider
• Search for a Higgs boson in the decay channel H to ZZ(*) to q qbar l-l+ in pp collisions at sqrt(s) = 7 TeV
• Challenges for Unsupervised Anomaly Detection in Particle Physics
• Online-compatible Unsupervised Non-resonant Anomaly Detection
• Deep Set Auto Encoders for Anomaly Detection in Particle Physics
• Transferability of Deep Learning Models in Searches for New Physics at Colliders