Speaker
Description
As the search for physics beyond the Standard Model widens, 'model-agnostic' searches, which do not assume any particular model of new physics, are increasing in importance. One promising model-agnostic search strategy is Classification Without Labels (CWoLa), in which a classifier is trained to distinguish events in a signal region from similar events in a sideband region, thereby learning about the presence of signal in the signal region. The CWoLa strategy was recently used in a full search for new physics in dijet events from Run-2 ATLAS data; in this search, only the masses of the two jets were used as classifier inputs. It has since been observed that while CWoLa performs well in such low-dimensional use cases, difficulties arise when adding additional jet features as classifier inputs. In this talk, we will describe ongoing work to combat these problems and extend the sensitivity of a CWoLa search by adding new observables to an ongoing analysis using $139$ $\text{fb}^{-1}$ of data from $pp$ collisions at $\sqrt{s}=$ 13 TeV in the ATLAS detector. In particular, we will discuss the anticipated benefits of adding classifier features, as well as the implementation of a simulation-assisted version of CWoLa which makes the strategy more robust.
| Are you are a member of the APS Division of Particles and Fields? | No |
|---|
