Seminar

Speaker

Yashar Akrami 

Département de Physique, École Normale Supérieure, Paris, France    

Title

Cosmic origins in light of Stage IV cosmological surveys, advanced numerical methods and cutting-edge inference techniques

Abstract

Our current standard models of particle physics and cosmology have provided us with a remarkably deep knowledge of the laws of physics that govern the Universe from the smallest to the largest scales and from the earliest to the latest times, but a plethora of important fundamental problems are yet to be addressed. In addition to the rapid development on the theoretical side over the past few decades, various large-scale cosmological collaborations have also been formed with the ambitious objective of mapping the entire observable Universe and answering a number of unsolved fundamental questions in cosmology and theoretical physics.

In this seminar, I will focus on the cosmology of the early Universe, and after a brief overview of our current understanding of cosmic origins and some of the open questions about the earliest moments in the cosmic history, I will discuss prospects for future efforts in this field of research. I will explain how some of the most fundamental questions about the primordial Universe are expected to be answered in the coming years using the wealth of cosmological data that will be provided by a large number of upcoming Stage IV surveys in combination with state-of-the-art inference and high-performance computing techniques and methods in artificial intelligence and data science. I will demonstrate this powerful synergistic science through some of my recent work on tests of models of the early Universe, as well as my leading contributions to the ongoing activities within a number of large-scale collaborations that aim to measure the cosmic microwave background and the cosmic large-scale structure with unprecedented precision.

Finally, I will introduce an ambitious, novel, powerful, synergistic, flexible, fast, efficient, accurate, automated, modular and end-to-end inference machinery, based on a combination of fast numerical simulations, ideas in artificial intelligence, and cutting-edge inference techniques, which I will develop in the coming years. This will reduce the computational cost by many orders of magnitude, making it possible to confront complex cosmological data with theoretical models in the era of precision cosmology. Consequently, it will allow us to obtain detailed information about the early Universe and cosmic initial conditions, and will significantly facilitate searches for new physics beyond the standard frameworks in cosmology and fundamental physics.

Questions from 11-11.15