Nicolosi, Pierpaolo
(2023)
A new emulator code for Cosmology: an application of machine learning algorithms to speed up cosmological analyses.
[Laurea magistrale], Università di Bologna, Corso di Studio in
Astrofisica e cosmologia [LM-DM270]
Documenti full-text disponibili:
![[img]](http://amslaurea.unibo.it/style/images/fileicons/application_pdf.png) |
Documento PDF (Thesis)
Disponibile con Licenza: Salvo eventuali più ampie autorizzazioni dell'autore, la tesi può essere liberamente consultata e può essere effettuato il salvataggio e la stampa di una copia per fini strettamente personali di studio, di ricerca e di insegnamento, con espresso divieto di qualunque utilizzo direttamente o indirettamente commerciale. Ogni altro diritto sul materiale è riservato
Download (3MB)
|
Abstract
In recent years, artificial intelligence has become omnipresent and is being used in various sectors, including cosmology. Machine learning algorithms are increasingly being used by scientific researchers to process and analyze the vast amounts of data provided by wide-field surveys and cosmological simulations. We have developed a machine learning-based code that speeds up cosmological analyses by emulating cosmological functions. The code is implemented in the public library CosmoBolognaLib and uses machine learning algorithms provided by the numerical library CosmoPower to build a neural network that imitates the output of the theoretical model of the two-point correlation function. We focused on emulating the model by varying four cosmological parameters and validated the emulator's accuracy by comparing its output to that of the original function. They applied their code to the analysis of cosmological simulations and performed a Bayesian analysis to derive the posterior probability distribution of the parameters, showing almost perfect correspondence with the original model. The innovative aspect of this work lies in the code's speed, which becomes thousands of times faster, reducing the total execution time from several tens of hours to a few seconds. Ww plan to improve the code by extending the emulation to other cosmological functions and expanding the emulator's range of validity to cover a wider parameter space. We believe that the potential applications of this methodology in the future are numerous and that its use will soon become a key element for future cosmological analyses conducted within the CosmoBolognaLib.
Abstract
In recent years, artificial intelligence has become omnipresent and is being used in various sectors, including cosmology. Machine learning algorithms are increasingly being used by scientific researchers to process and analyze the vast amounts of data provided by wide-field surveys and cosmological simulations. We have developed a machine learning-based code that speeds up cosmological analyses by emulating cosmological functions. The code is implemented in the public library CosmoBolognaLib and uses machine learning algorithms provided by the numerical library CosmoPower to build a neural network that imitates the output of the theoretical model of the two-point correlation function. We focused on emulating the model by varying four cosmological parameters and validated the emulator's accuracy by comparing its output to that of the original function. They applied their code to the analysis of cosmological simulations and performed a Bayesian analysis to derive the posterior probability distribution of the parameters, showing almost perfect correspondence with the original model. The innovative aspect of this work lies in the code's speed, which becomes thousands of times faster, reducing the total execution time from several tens of hours to a few seconds. Ww plan to improve the code by extending the emulation to other cosmological functions and expanding the emulator's range of validity to cover a wider parameter space. We believe that the potential applications of this methodology in the future are numerous and that its use will soon become a key element for future cosmological analyses conducted within the CosmoBolognaLib.
Tipologia del documento
Tesi di laurea
(Laurea magistrale)
Autore della tesi
Nicolosi, Pierpaolo
Relatore della tesi
Correlatore della tesi
Scuola
Corso di studio
Ordinamento Cds
DM270
Parole chiave
machine learning emulator speed up cosmological analysis
Data di discussione della Tesi
26 Maggio 2023
URI
Altri metadati
Tipologia del documento
Tesi di laurea
(NON SPECIFICATO)
Autore della tesi
Nicolosi, Pierpaolo
Relatore della tesi
Correlatore della tesi
Scuola
Corso di studio
Ordinamento Cds
DM270
Parole chiave
machine learning emulator speed up cosmological analysis
Data di discussione della Tesi
26 Maggio 2023
URI
Statistica sui download
Gestione del documento: