FETA: Flow-Enhanced Transportation for Anomaly Detection

• DOI: 10.48550/arxiv.2212.11285
• Publication Date: 2022-01-01
• Authors: T. Golling; Samuel Klein; Radha Mastandrea; Benjamin Nachman
• Keywords: Anomaly (physics); SIGNAL (programming language); Feature (linguistics)

Resonant anomaly detection is a promising framework for model-independent searches new particles. Weakly supervised resonant methods compare data with potential signal against template of the Standard Model (SM) background inferred from sideband regions. We propose means to generate this that uses flow-based model create mapping between high-fidelity SM simulations and data. The flow trained in regions region blinded, conditioned on feature (mass) such it can be interpolated into region. To illustrate approach, we use simulated collisions Large Hadron Collider (LHC) Olympics Dataset. find our flow-constructed method has competitive sensitivity other recent proposals therefore provide complementary information improve future searches.