A Normalized Autoencoder for LHC Triggers

Dillon, Barry M.; Favaro, L; Plehn, Tilman; Sorrenson, Peter; Krämer, M.

doi:10.48550/arxiv.2206.14225

Cited by 6 publications

(5 citation statements)

References 76 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While this is not strictly density estimation, the optimisation is highly aligned with learning a density, since regions of the phase space which are most populated are those which should be reconstructed the best and thus have the lowest anomaly score. There has been significant progress with the AutoEncoder tools and other density-based anomaly detection methods in recent years [19][20][21][22][23][24][25][26][27][28][29][30][31][32][33], with studies covering interpretability of AutoEncoders [34,35], topic modelling [36,37], null hypothesis tests for anomaly detection [38], ABCD methods [39], the Normalised AutoEncoder (NAE) [40], and normalising flow techniques [41][42][43][44]. For a comprehensive summary of many different anomaly detection methods we refer the reader to the community challenge papers in Refs [45,46].…”

Section: Introductionmentioning

confidence: 99%

Anomalies, Representations, and Self-Supervision

Dillon¹,

Favaro²,

Feiden³

et al. 2023

Preprint

View full text Add to dashboard Cite

We develop a self-supervised method for density-based anomaly detection using contrastive learning, and test it using event-level anomaly data from CMS ADC2021. The Anomaly-CLR technique is data-driven and uses augmentations of the background data to mimic non-Standard-Model events in a model-agnostic way. It uses a permutation-invariant Transformer Encoder architecture to map the objects measured in a collider event to the representation space, where the data augmentations define a representation space which is sensitive to potential anomalous features. An AutoEncoder trained on background representations then computes anomaly scores for a variety of signals in the representation space. With AnomalyCLR we find significant improvements on performance metrics for all signals when compared to the raw data baseline.

show abstract

Section: Introductionmentioning

confidence: 99%

Anomalies, Representations, and Self-Supervision

Dillon¹,

Favaro²,

Feiden³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…However, they are of a different nature than most current implementations of generative networks which are based on variations of Autoencoders (see e.g. [42][43][44][45][46]), Generative Adversarial Networks (see e.g. [44,47]), Normalizing Flows (see e.g.…”

Section: Introductionmentioning

confidence: 99%

Exploring unsupervised top tagging using Bayesian inference

et al. 2023

View full text Add to dashboard Cite

Recognizing hadronically decaying top-quark jets in a sample of jets, or even its total fraction in the sample, is an important step in many LHC searches for Standard Model and Beyond Standard Model physics as well. Although there exists outstanding top-tagger algorithms, their construction and their expected performance rely on Montecarlo simulations, which may induce potential biases. For these reasons we develop two simple unsupervised top-tagger algorithms based on performing Bayesian inference on a mixture model. In one of them we use as the observed variable a new geometrically-based observable \tilde{A}_{3}Ã3, and in the other we consider the more traditional \tau_{3}/\tau_{2}τ3/τ2NN-subjettiness ratio, which yields a better performance. As expected, we find that the unsupervised tagger performance is below existing supervised taggers, reaching expected Area Under Curve AUC \sim 0.80-0.81∼0.80−0.81 and accuracies of about 69% -− 75% in a full range of sample purity. However, these performances are more robust to possible biases in the Montecarlo that their supervised counterparts. Our findings are a step towards exploring and considering simpler and unbiased taggers.

show abstract

“…In parallel, the rapid development of machine learning to the analysis of jet energy depositions [19,22,27] demonstrated that jet tagging strategies, including those for semivisible jets, can be learned directly from lower-level jet constituents without the need to form physics-motivated high level observables [17,28]. Such learned models are naturally challenging to interpret, validate or quantify uncertainties, especially given the highdimensional nature of their inputs.…”

Section: Jhep12(2022)132 1 Introductionmentioning

confidence: 99%

Learning to identify semi-visible jets

Faucett

Hsu

Whiteson

2022

J. High Energ. Phys.

View full text Add to dashboard Cite

We train a network to identify jets with fractional dark decay (semi-visible jets) using the pattern of their low-level jet constituents, and explore the nature of the information used by the network by mapping it to a space of jet substructure observables. Semi-visible jets arise from dark matter particles which decay into a mixture of dark sector (invisible) and Standard Model (visible) particles. Such objects are challenging to identify due to the complex nature of jets and the alignment of the momentum imbalance from the dark particles with the jet axis, but such jets do not yet benefit from the construction of dedicated theoretically-motivated jet substructure observables. A deep network operating on jet constituents is used as a probe of the available information and indicates that classification power not captured by current high-level observables arises primarily from low-pT jet constituents.

show abstract

A Normalized Autoencoder for LHC Triggers

Cited by 6 publications

References 76 publications

Anomalies, Representations, and Self-Supervision

Anomalies, Representations, and Self-Supervision

Exploring unsupervised top tagging using Bayesian inference

Learning to identify semi-visible jets

Contact Info

Product

Resources

About