`

Timezone: »

 
Classifying Anomalies THrough Outer Density Estimation (CATHODE)
Joshua Isaacson · Gregor Kasieczka · Benjamin Nachman · David Shih · Manuel Sommerhalder

We propose a new model-agnostic search strategy for hints of new fundamental forces motivated by applications in particle physics. It is based on a novel application of neural density estimation to anomaly detection. Our approach, which we call Classifying Anomalies THrough Outer Density Estimation (CATHODE), assumes potential signal events cluster in phase space in a signal region. However, backgrounds due to known processes are also present in the signal region and too large to directly detect such a signal. By training a conditional density estimator on a collection of additional features outside the signal region, interpolating it into the signal region, and sampling from it, we produce a collection of events that follow the background model. We can then train a classifier to distinguish the data from the events sampled from the background model, thereby approaching the optimal anomaly detector. Using the public LHC Olympics R&D data set, we demonstrate that CATHODE nearly saturates the best possible performance, and significantly outperforms other approaches in this bump hunt paradigm.

Author Information

Joshua Isaacson (Fermi National Accelerator Laboratory)
Gregor Kasieczka (Universität Hamburg)
Benjamin Nachman (Lawrence Berkeley National Laboratory)
David Shih ( Rutgers University )
Manuel Sommerhalder (Universität Hamburg)

More from the Same Authors