A guide to constraining effective field theories with machine learning

By J. Brehmer, Kyle Cranmer, Gilles Louppe, Juan Pavez. 2018

157 citations recorded by the source corpus.

A proposed machine-learning approach could speed up the analysis that underlies searches for new particles in high-energy collisions.

Subfields: Nuclear and High Energy Physics, AI for Physical Sciences

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Nuclear and High Energy Physics

In AI for Physical Sciences

Citations in the corpus

The frontier of simulation-based inference (2019, methodology): “Other diagnostic tools that can be applied at the end of the inference stage involve training classifiers to distinguish data from the surrogate model and the true simulator (60), self-consistency checks (9, 60), ensemble methods, and comparing distributions of network output against known asymptoti”
Likelihood-free MCMC with Approximate Likelihood Ratios (2019, background): “However, recent works [57, 58] make it possible to signiﬁcantly reduce the cost of this step, provided joint likelihood ratios and scores can be extracted from the simulator.”
Mining for Dark Matter Substructure: Inferring Subhalo Population Properties from Strong Lenses with Machine Learning (2019, methodology): “In this paper, we follow this approach and apply a particularly powerful technique for simulation-based inference introduced in Brehmer et al. (2018a, 2018b, 2018c) and Stoye et al. (2018) to the problem of extracting high-level substructure properties from an ensemble of galaxy–galaxy…”
Likelihood-free MCMC with Amortized Approximate Ratio Estimators (2019, background): “Recent works (Brehmer et al., 2020; 2018) signiﬁcantly reduce the amount of required simulations, provided joint likelihood ratios and scores can be extracted from the simulator.”
Efficient Probabilistic Inference in the Quest for Physics Beyond the Standard Model (2018, background): “Recently, machine learning techniques have been employed to learn surrogates for the implicit densities defined by the simulators as a strategy for likelihood-free inference [59].”
Likelihood-free inference with an improved cross-entropy estimator (2018, methodology): “[25].”
INFERNO: Inference-Aware Neural Optimisation (2018, background): “[18, 19, 20] further extended the approach of parametrised classifiers to better exploit the latent-space space structure of generative models from complex scientific simulators.”
Energy flow networks: deep sets for particle jets (2018, unknown): “Additional tasks include the removal of pileup [30], model-independent new physics searches [31–36], constraining effective field theories [37–39], probabilistic and generative modeling of physics processes [40–44], and enhancing existing physics analyses [45–53].”
Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows (2018, methodology): “For instance, inferring the parameters of physical theories given particle collisions can be the means for discovering new physics [11], and inferring the scene parameters of a graphics engine given a natural image can be the basis for computer vision [66, 87].”
Fast likelihood-free cosmology with neural density estimators and active learning (2019, methodology): “As a third approach, the score-function may be regressed directly from simulations in a likelihood-free manner using neural networks (Brehmer et al. 2018a,b,c).”