Abstract
The lack of evidence for new physics at the Large Hadron Collider so far has prompted the development of model-independent search techniques. In this study, we compare the anomaly scores of a variety of anomaly detection techniques: an isolation forest, a Gaussian mixture model, a static autoencoder, and a β-variational autoencoder (VAE), where we define the reconstruction loss of the latter as a weighted combination of regression and classification terms. We apply these algorithms to the 4-vectors of simulated LHC data, but also investigate the performance when the non-VAE algorithms are applied to the latent space variables created by the VAE. In addition, we assess the performance when the anomaly scores of these algorithms are combined in various ways. Using super- symmetric benchmark points, we find that the logical AND combination of the anomaly scores yielded from algorithms trained in the latent space of the VAE is the most effective discriminator of all methods tested.
Highlights
At the LHC and the Tevatron [16]
We find that the logical AND combination of the anomaly scores yielded from algorithms trained in the latent space of the variational autoencoder (VAE) is the most effective discriminator of all methods tested
We investigate how the performance changes when the non-VAE techniques are run on the latent space variables of the VAE, supplemented by the VAE reconstruction loss
Summary
At the LHC and the Tevatron [16]. Other model-independent LHC searches have been performed by the ATLAS and CMS collaborations, or with their publicly available data [17,18,19,20,21]. In this paper we perform a systematic comparison of a variety of anomaly detection techniques for LHC searches, including novel ideas such as training within the latent space of a variational autoencoder, and combining algorithms in various ways. The scope of our paper is to provide a proof of principle that our techniques are viable for some realistic new physics scenarios In this regard, we note that some of the supersymmetric scenarios are already comfortably excluded by current dedicated ATLAS and CMS searches. Our proposed techniques are designed to provide a model-independent approach to LHC searches, it is useful to test their performance on particular models of interest. We use a variety of supersymmetric benchmark models to test our techniques, using the supersymmetric signal and SM background processes from the dataset published and described in ref. [38]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
