Search for low mass dark matter in DarkSide-50: the bayesian network approach

Agnes P.;Albuquerque I. F. M.;Alexander T.;Alton A. K.;Ave M.;Back H. O.;Batignani G.;Biery K.;Bocci V.;Bonivento W. M.;Bottino B.;Bussino S.;Cadeddu M.;Cadoni M.;Calaprice F.;Caminata A.;Campos M. D.;Canci N.;Caravati M.;Cargioli N.;Cariello M.;Carlini M.;Cataudella V.;Cavalcante P.;Cavuoti S.;Chashin S.;Chepurnov A.;Cicalo C.;Covone G.;D'Angelo D.;Davini S.;De Candia A.;De Cecco S.;De Filippis G.;De Rosa G.;Derbin A. V.;Devoto A.;D'Incecco M.;Dionisi C.;Dordei F.;Downing M.;D'Urso D.;Fairbairn M.;Fiorillo G.;Franco D.;Gabriele F.;Galbiati C.;Ghiano C.;Giganti C.;Giovanetti G. K.;Goretti A. M.;Grilli di Cortona G.;Grobov A.;Gromov M.;Guan M.;Gulino M.;Hackett B. R.;Herner K.;Hessel T.;Hosseini B.;Hubaut F.;Hungerford E. V.;Ianni A.;Ippolito V.;Keeter K.;Kendziora C. L.;Kimura M.;Kochanek I.;Korablev D.;Korga G.;Kubankin A.;Kuss M.;La Commara M.;Lai M.;Li X.;Lissia M.;Longo G.;Lychagina O.;Machulin I. N.;Mapelli L. P.;Mari S. M.;Maricic J.;Messina A.;Milincic R.;Monroe J.;Morrocchi M.;Mougeot X.;Muratova V. N.;Musico P.;Nozdrina A. O.;Oleinik A.;Ortica F.;Pagani L.;Pallavicini M.;Pandola L.;Pantic E.;Paoloni E.;Pelczar K.;Pelliccia N.;Piacentini S.;Pocar A.;Poehlmann D. M.;Pordes S.;Poudel S. S.;Pralavorio P.;Price D. D.;Ragusa F.;Razeti M.;Razeto A.;Renshaw A. L.;Rescigno M.;Rode J.;Romani A.;Sablone D.;Samoylov O.;Sandford E.;Sands W.;Sanfilippo S.;Savarese C.;Schlitzer B.;Semenov D. A.;Shchagin A.;Sheshukov A.;Skorokhvatov M. D.;Smirnov O.;Sotnikov A.;Stracka S.;Suvorov Y.;Tartaglia R.;Testera G.;Tonazzo A.;Unzhakov E. V.;Vishneva A.;Vogelaar R. B.;Wada M.;Wang H.;Wang Y.;Westerdale S.;Wojcik M. M.;Xiao X.;Yang C.;Zuzel G.

2023-01-01

---

Abstract

We present a novel approach for the search of dark matter in the DarkSide-50 experiment, relying on Bayesian Networks. This method incorporates the detector response model into the likelihood function, explicitly maintaining the connection with the quantity of interest. No assumptions about the linearity of the problem or the shape of the probability distribution functions are required, and there is no need to morph signal and background spectra as a function of nuisance parameters. By expressing the problem in terms of Bayesian Networks, we have developed an inference algorithm based on a Markov Chain Monte Carlo to calculate the posterior probability. A clever description of the detector response model in terms of parametric matrices allows us to study the impact of systematic variations of any parameter on the final results. Our approach not only provides the desired information on the parameter of interest, but also potential constraints on the response model. Our results are consistent with recent published analyses and further refine the parameters of the detector response model.