Abstract: Optimisation problems are ubiquitous in particle and astrophysics, and involve locating the optimum of a complicated function of many parameters that may be computationally expensive to evaluate. We describe a number of global optimisation algorithms that are not yet widely used in particle astrophysics, benchmark them against random sampling and existing techniques, and perform a detailed comparison of their performance on a range of test functions. These include four analytic test functions of varying dimensionality, and a realistic example derived from a recent global fit of weak-scale supersymmetry. Although the best algorithm to use depends on the function being investigated, we are able to present general conclusions about the relative merits of random sampling, Differential Evolution, Particle Swarm Optimisation, the Covariance Matrix Adaptation Evolution Strategy, Bayesian Optimisation, Grey Wolf Optimisation, and the PyGMO Artificial Bee Colony, Gaussian Particle Filter and Adaptive Memory Programming for Global Optimisation algorithms.

Abstract: We assess the coverage properties of confidence and credible intervals on the CMSSM parameter space inferred from a Bayesian posterior and the profile likelihood based on an ATLAS sensitivity study. In order to make those calculations feasible, we introduce a new method based on neural networks to approximate the mapping between CMSSM parameters and weak-scale particle masses. Our method reduces the computational effort needed to sample the CMSSM parameter space by a factor of similar to 10(4) with respect to conventional techniques. We find that both the Bayesian posterior and the profile likelihood intervals can significantly over-cover and identify the origin of this effect to physical boundaries in the parameter space. Finally, we point out that the effects intrinsic to the statistical procedure are conflated with simplifications to the likelihood functions from the experiments themselves.

Abstract: Searches for dark matter produced via scalar resonances in final states consisting of Standard Model (SM) particles and missing transverse momentum are of high relevance at the LHC. Motivated by dark-matter portal models, most existing searches are optimized for unbalanced decay topologies for which the missing momentum recoils against the visible SM particles. In this work, we show that existing searches are also sensitive to a wider class of models, which we characterize by a recently presented simplified model framework. We point out that searches for models with a balanced decay topology can be further improved with more dedicated analysis strategies. For this study, we investigate the feasibility of a new search for bottom-quark associated neutral Higgs production with a b (b) over barZ + p(T)(miss) final state and perform a detailed collider analysis. Our projected results in the different simplified model topologies investigated here can be easily reinterpreted in a wide range of models of physics beyond the SM, which we explicitly demonstrate for the example of the Two-Higgs-Doublet model with an additional pseudoscalar Higgs boson.

Abstract: The framed Standard Model (FSM) predicts a 0(+) boson with mass around 20 MeV in the “hidden sector,” which mixes at tree level with the standard Higgs hW and hence acquires small couplings to quarks and leptons which can be calculated in the FSM apart from the mixing parameter rho Uh. The exchange of this mixed state U will contribute to g – 2 and to the Lamb shift. By adjusting rho Uh alone, it is found that the FSM can satisfy all present experimental bounds on the g – 2 and Lamb shift anomalies for μand e, and for the latter for both hydrogen and deuterium. The FSM predicts also a 1(-) boson in the “hidden sector” with a mass of 17 MeV, that is, right on top of the Atomki anomaly X. This mixes with the photon at 1-loop level and couples thereby like a dark photon to quarks and leptons. It is however a compound state and is thought likely to possess additional compound couplings to hadrons. By adjusting the mixing parameter and the X's compound coupling to nucleons, the FSM can reproduce the production rate of the X in beryllium decay as well as satisfy all the bounds on X listed so far in the literature. The above two results are consistent in that the U, being 0(+), does not contribute to the Atomki anomaly if parity and angular momentum are conserved, while X, though contributing to g – 2 and Lamb shift, has smaller couplings than U and can, at first instance, be neglected there. Thus, despite the tentative nature of the three anomalies in experiment on the one hand and of the FSM as theory on the other, the accommodation of the former in the latter has strengthened the credibility of both. Indeed, if this FSM interpretation were correct, it would change the whole aspect of the anomalies from just curiosities to windows into a vast hitherto hidden sector comprising at least in part the dark matter which makes up the bulk of our universe.

Abstract: Statistical inference of the fundamental parameters of supersymmetric theories is a challenging and active endeavor. Several sophisticated algorithms have been employed to this end. While Markov-Chain Monte Carlo (MCMC) and nested sampling techniques are geared towards Bayesian inference, they have also been used to estimate frequentist confidence intervals based on the profile likelihood ratio. We investigate the performance and appropriate configuration of MULTINEST, a nested sampling based algorithm, when used for profile likelihood-based analyses both on toy models and on the parameter space of the Constrained MSSM. We find that while the standard configuration previously used in the literarture is appropriate for an accurate reconstruction of the Bayesian posterior, the profile likelihood is poorly approximated. We identify a more appropriate MULTINEST configuration for profile likelihood analyses, which gives an excellent exploration of the profile likelihood (albeit at a larger computational cost), including the identification of the global maximum likelihood value. We conclude that with the appropriate configuration MULTINEST is a suitable tool for profile likelihood studies, indicating previous claims to the contrary are not well founded.