Exploring the BSM parameter space with Neural Network aided Simulation-Based Inference

Abstract

Some of the issues that make sampling parameter spaces of various beyond the Standard Model (BSM) scenarios computationally expensive are the high dimensionality of the input parameter space, complex likelihoods, and stringent experimental constraints. In this work, we explore likelihood-free approaches, leveraging neural network-aided Simulation-Based Inference (SBI) to alleviate this issue. We focus on three amortized SBI methods: Neural Posterior Estimation (NPE), Neural Likelihood Estimation (NLE), and Neural Ratio Estimation (NRE) and perform a comparative analysis through the validation test known as the \textit{ Test of Accuracy with Random Points} (TARP), as well as through posterior sample efficiency and computational time. As an example, we focus on the scalar sector of the phenomenological minimal supersymmetric SM (pMSSM) and observe that the NPE method outperforms the others and generates correct posterior distributions of the parameters with a minimal number of samples. The efficacy of this framework is tested on 5 parameter pMSSM with Higgs and flavor physics data and its performance is compared with the MCMC method. We further add dark matter (DM) observables to make the task more challenging and consider a 9 parameter pMSSM. We observe that even though the efficiency factor drops, the amortized SBI method still produces faithful posterior distributions. SBI predicted points satisfying DM constraints are mostly bino-dominated upto $\sim$ 1.5 TeV, and are mostly wino-dominated within the 1.5 - 2 TeV range.

Accepted by JHEP for publication

References in corpus (49)

• Theory and phenomenology of two-Higgs-doublet models
• The Anatomy of Electro-Weak Symmetry Breaking. II: The Higgs bosons in the Minimal Supersymmetric Model
• micrOMEGAs3.1 : a program for calculating dark matter observables
• Dark matter direct detection rate in a generic model with micrOMEGAs2.2
• Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification
• 2020 Global reassessment of the neutrino oscillation picture
• micrOMEGAs: Version 1.3
• Indirect search for dark matter with micrOMEGAs2.4
• Precise determination of the neutral Higgs boson masses in the MSSM
• Precise prediction for the light MSSM Higgs boson mass combining effective field theory and fixed-order calculations
• Likelihood Analysis of the pMSSM11 in Light of LHC 13-TeV Data
• Precision calculations in the MSSM Higgs-boson sector with FeynHiggs 2.14
• MSSM Higgs Bosons at The LHC
• A global fit of the MSSM with GAMBIT
• Full one-loop corrections to the relic density in the MSSM: A few examples
• Recasting direct detection limits within micrOMEGAs and implication for non-standard Dark Matter scenarios
• Direct Detection of the Wino- and Higgsino-like Neutralino Dark Matters at One-Loop Level
• Prospects for dark matter searches in the pMSSM
• Dark matter and Higgs bosons in the MSSM
• Relic density at one-loop with gauge boson pair production
• The Electroweak Sector of the pMSSM in the Light of LHC - 8 TeV and Other Data
• Reduced LHC constraints for higgsino-like heavier electroweakinos
• Deep learning in the heterotic orbifold landscape
• Many faces of low mass neutralino dark matter in the unconstrained MSSM, LHC data and new signals
• How light a higgsino or a wino dark matter can become in a compressed scenario of MSSM
• LtU-ILI: An All-in-One Framework for Implicit Inference in Astrophysics and Cosmology
• Implications of the Higgs discovery for the MSSM
• Constraining the Parameters of High-Dimensional Models with Active Learning
• Low mass neutralino dark matter in mSUGRA and more general models in the light of LHC data
• Neutralino dark matter confronted by the LHC constraints on Electroweak SUSY signals
• Status of MSSM Higgs Sector using Global Analysis and Direct Search Bounds, and Future Prospects at the HL-LHC
• Efficient sampling of constrained high-dimensional theoretical spaces with machine learning
• Probing the light Higgs pole resonance annihilation of dark matter in the light of XENON100 and CDMS-II observations
• Current status of MSSM Higgs sector with LHC 13 TeV data
• Exploration of Parameter Spaces Assisted by Machine Learning
• Measurement of off-shell Higgs boson production in the $H^{\ast }\to ZZ\to 4\ell$ decay channel using a neural simulation-based inference technique in 13 TeV $pp$ collisions with the ATLAS detector
• An implementation of neural simulation-based inference for parameter estimation in ATLAS
• Active learning BSM parameter spaces
• Combining Evolutionary Strategies and Novelty Detection to go Beyond the Alignment Limit of the $Z_3$ 3HDM
• Constraining the Higgs Potential with Neural Simulation-based Inference for Di-Higgs Production
• Advancing Tools for Simulation-Based Inference
• Simulation Based Inference for Efficient Theory Space Sampling: an Application to Supersymmetric Explanations of the Anomalous Muon (g-2)
• Current status of the light neutralino thermal dark matter in the phenomenological MSSM
• Reconstructing axion-like particles from beam dumps with simulation-based inference
• Bayesian Active Search on Parameter Space: a 95 GeV Spin-0 Resonance in the ($B-L$)SSM
• Fast multilabel classification of HEP constraints with deep learning
• Machine-Learning the Classification of Spacetimes
• Measurements of charmonia decays from BESIII
• Data-Driven High-Dimensional Statistical Inference with Generative Models