Machine learning models for the ATLAS dark matter trigger

Dark matter is invisible and therefore does not interact with the ATLAS detector. To detect and save these events with the trigger system, the missing transverse momentum (MET) can be reconstructed. This poster will provide an overview of the ATLAS trigger, including new methods implemented in the last year of data taking. I will then explore a neural network to optimize the ability of the network to differentiate between background and signal data and to estimate the MET in the event using a regression.