Anomaly Detection using Autoencoders on Fundamental LZ Signals

Searching for anomalous data is especially important in rare event searches like that of the LUX-ZEPLIN (LZ) experiment’s hunt for dark matter. While LZ’s data processing provides analyzer-friendly features for all data, searching for anomalous data after minimal reconstruction allows one to find anomalies not captured by reconstructed features, agnostic to any reconstruction errors. Autoencoders can be used to probe for anomalous light-detecting photomultiplier tube (PMT) waveforms resulting from ionization signals (S2) and have found unresolved S2s resulting from multiple scatter interactions. These techniques can be extended by applying them to prompt scintillation light (S1) PMT waveforms and by analyzing the latent-spaces of variational autoencoders (VAEs).