Unsupervised outlier detection in heavy-ion collisions

We present different methods of unsupervised learning which can be used for outlier detection in high energy nuclear collisions. This method is of particular interest for heavy ion collisions where a direct comparison of experimental data to model simulations is often ambiguous and it is not easy to determine whether an observation is due to new physics, an incomplete understanding of the known physics or an experimental artefact. The UrQMD model is used to generate the bulk background of events as well as different variants of outlier events which may result from misidentified centrality or detector malfunctions. The methods presented here can be generalized to different and novel physics effects. To detect the outliers, dimensional reduction algorithms are implemented, speciftically the Principle Component Analysis (PCA) and Autoencoders (AEN). We find that mainly the reconstruction error is a good measure to distinguish outliers from background. The performance of the algorithms is compared using a ROC curve. It is shown that the number of reduced (encoded) dimensions to describe a single event contributes significantly to the performance of the outlier detection task. We find that the model which is best suited to separate outlier events requires a good performance in reconstructing events and at the same time a small number of parameters.

Automated summary

This study presents various unsupervised learning methods for outlier detection in high energy nuclear collisions, especially in heavy ion collisions. Dimensional reduction algorithms like PCA and AEN are used to distinguish outliers from background. The research finds that the most effective model for separating outlier events requires good performance in event reconstruction while maintaining a small number of parameters.