While deep learning has proven to be extremely successful at supervised classification tasks at the LHC and beyond, for practical applications, raw classification accuracy is often not the only ...consideration. One crucial issue is the stability of network predictions, either versus changes of individual features of the input data or against systematic perturbations. We present a new method based on a novel application of "distance correlation," a measure quantifying nonlinear correlations, that achieves equal performance to state-of-the-art adversarial decorrelation networks but is much simpler and more stable to train. To demonstrate the effectiveness of our method, we carefully recast a recent ATLAS study of decorrelation methods as applied to boosted, hadronic W tagging. We also show the feasibility of regularization with distance correlation for more powerful convolutional neural networks, as well as for the problem of hadronic top tagging.
Deep-learning top taggers or the end of QCD? Kasieczka, Gregor; Plehn, Tilman; Russell, Michael ...
The journal of high energy physics,
05/2017, Volume:
2017, Issue:
5
Journal Article
Peer reviewed
Open access
A
bstract
Machine learning based on convolutional neural networks can be used to study jet images from the LHC. Top tagging in fat jets offers a well-defined framework to establish our DeepTop ...approach and compare its performance to QCD-based top taggers. We first optimize a network architecture to identify top quarks in Monte Carlo simulations of the Standard Model production channel. Using standard fat jets we then compare its performance to a multivariate QCD-based top tagger. We find that both approaches lead to comparable performance, establishing convolutional networks as a promising new approach for multivariate hypothesis-based top tagging.
Autoencoder networks, trained only on QCD jets, can be used to search
for anomalies in jet-substructure. We show how, based either on images
or on 4-vectors, they identify jets from decays of ...arbitrary heavy
resonances. To control the backgrounds and the underlying systematics we
can de-correlate the jet mass using an adversarial network. Such an
adversarial autoencoder allows for a general and at the same time easily
controllable search for new physics. Ideally, it can be trained and
applied to data in the same phase space region, allowing us to
efficiently search for new physics using un-supervised learning.
A
bstract
The past few years have seen a rapid development of machine-learning algorithms. While surely augmenting performance, these complex tools are often treated as black-boxes and may impair our ...understanding of the physical processes under study. The aim of this paper is to move a first step into the direction of applying expert-knowledge in particle physics to calculate the optimal decision function and test whether it is achieved by standard training, thus making the aforementioned black-box more transparent. In particular, we consider the binary classification problem of discriminating quark-initiated jets from gluon-initiated ones. We construct a new version of the widely used
N
-subjettiness, which features a simpler theoretical behaviour than the original one, while maintaining, if not exceeding, the discrimination power. We input these new observables to the simplest possible neural network, i.e. the one made by a single neuron, or perceptron, and we analytically study the network behaviour at leading logarithmic accuracy. We are able to determine under which circumstances the perceptron achieves optimal performance. We also compare our analytic findings to an actual implementation of a perceptron and to a more realistic neural network and find very good agreement.
Unsupervised hadronic SUEP at the LHC Barron, Jared; Curtin, David; Kasieczka, Gregor ...
The journal of high energy physics,
12/2021, Volume:
2021, Issue:
12
Journal Article
Peer reviewed
Open access
A
bstract
Confining dark sectors with pseudo-conformal dynamics produce SUEPs, or Soft Unclustered Energy Patterns, at colliders: isotropic dark hadrons with soft and democratic energies. We target ...the experimental nightmare scenario, SUEPs in exotic Higgs decays, where all dark hadrons decay promptly to SM hadrons. First, we identify three promising observables: the charged particle multiplicity, the event ring isotropy, and the matrix of geometric distances between charged tracks. Their patterns can be exploited through a cut-and-count search, supervised machine learning, or an unsupervised autoencoder. We find that the HL-LHC will probe exotic Higgs branching ratios at the per-cent level, even without a detailed knowledge of the signal features. Our techniques can be applied to other SUEP searches, especially the unsupervised strategy, which is independent of overly specific model assumptions and the corresponding precision simulations.
We introduce a new and highly efficient tagger for hadronically decaying top quarks, based on a deep neural network working with Lorentz vectors and the Minkowski metric. With its novel machine ...learning setup and architecture it allows us to identify boosted top quarks not only from calorimeter towers, but also including tracking information. We show how the performance of our tagger compares with QCD-inspired and image-recognition approaches and find that it significantly increases the performance for strongly boosted top quarks.
Distinguishing quarks from gluons based on low-level detector output
is one of the most challenging applications of multi-variate and machine
learning techniques at the LHC. We first show the ...performance of our
4-vector-based LoLa tagger without and after considering detector
effects. We then discuss two benchmark applications, mono-jet searches
with a gluon-rich signal and di-jet resonances with a quark-rich signal.
In both cases an immediate benefit compared to the standard event-level
analysis exists.
How to GAN away detector effects Bellagente, Marco; Butter, Anja; Kasieczka, Gregor ...
SciPost physics,
04/2020, Volume:
8, Issue:
4
Journal Article
Peer reviewed
Open access
LHC analyses directly comparing data and simulated events bear the
danger of using first-principle predictions only as a black-box
part of event simulation. We show how simulations, for instance, of
...detector effects can instead be inverted using generative
networks. This allows us to reconstruct parton level information
from measured events. Our results illustrate how, in general, fully
conditional generative networks can statistically invert Monte Carlo
simulations. As a technical by-product we show how a maximum mean
discrepancy loss can be staggered or cooled.
A critical question concerning generative networks applied to event generation in particle physics is if the generated events add statistical precision beyond the training sample. We show for a ...simple example with increasing dimensionality how generative networks indeed amplify the training statistics. We quantify their impact through an amplification factor or equivalent numbers of sampled events.
We show how to treat systematic uncertainties using Bayesian deep
networks for regression. First, we analyze how these networks
separately trace statistical and systematic uncertainties on the
...momenta of boosted top quarks forming fat jets. Next, we propose a
novel calibration procedure by training on labels and their error bars.
Again, the network cleanly separates the different uncertainties. As
a technical side effect, we show how Bayesian networks can be extended to
describe non-Gaussian features.