Modeling a Theory of Gravity Using Machine Learning



In this paper, we study graviton interactions, a key part of modelling a theory of gravity. Since no numerical calculate schemes have been put forward to calculate functions that describe these interactions, we use general principles of quantum mechanics to come up with constraints which bound these interaction functions, a technique known as Bootstrapping. We use Neural Networks to generate functions that satisfy those constraints, for which no explicit example currently exists. Limiting our study to 2-2 gravitons, we show that Neural Networks can successfully model quantum scattering interactions, producing physically reasonable results. However, we find that the final output function is still dependent on the initial conditions, indicating that our model has not yet found an optimal solution. Despite this, our scheme still provides the first proof of concept of using machine learning for theoretically analyzing quantum gravity.

Speaker: Mohammad Saad Naeem is doing his Undergraduate degree from McGill University (Montreal, Canada) in Physics and Computer Science. He is very passionate about using computers as a tool to understand some of the most fundamental problems in physics. He finds Quantum Computing fascinating because of the competitive aspect of achieving quantum supremacy and the challenging physics involved.
Saad has research experience in different areas of quantum computing such as photonic quantum networks, error correction on rotationally invariant cores, and has worked this summer as a quantum software engineer with Artiste QB. After graduating this December, he hopes to continue his journey as the next generation of scientists in the quantum computing industry.

Moderators: Pawel Gora (CEO of Quantum AI Foundation), Aroosa Ijaz (phD candidate of University of Waterloo), Menna El-Masry (researcher of Alexandria University)

The event is finished.


Nov 06 2021


1:00 pm - 3:00 pm

Leave A Reply

Your email address will not be published.