QUC School on 

  A.I. in High Energy Physics

 

 

 

 

 

 

 

 

 

 

 

 

 

 

       JULY 20 - 31, 2020 l Online                                                                                                                                                  

Colloquium & Seminar Home > Colloquium & Seminar

Colloquium

 

· Speaker:  Matthew Schwartz 

· Date & time: Friday, July 24, 9:30-10:30 

· Title:  Modern Machine Learning and the Large Hadron Collider 

· Abstract: Particle physics involves an intense data science challenge: signals of interest, like for the Higgs boson are often buried in backgrounds trillions of times as large. Over the last decade or so, there has been a revolution in machine learning, and over the last 5 years of so, the revolution has started to influence collider physics. For example, image recognition technology based on convolutional neural networks can be used to find patterns in the distribution of energy around a detector. Or recurrent neural networks designed for natural language processing can be applied to parse the substructure of jets. This talk will review some of these developments and discuss some exciting possibilities for the future. 

· Zoom link:  https://zoom.us/j/99902265746?pwd=aGlVNkI5TWZpSHl6QW9tMjV0ejJaUT09

 

 

 

Seminar

 

· Speaker:  Michael Spannowsky(IPPP, Durham U)

· Date & time: Wednesday, July 29, 4:30-5:30pm

· Title:  Novel (Quantum) Computational Methods for Quantum Field Theories

· Abstract:  I will discuss novel approaches to for the task of finding a solution to a quantum field theoretical problem, e.g. tunnelling, in terms of an optimisation problem that can be solved either classically using machine learning methods or through a quantum computational ansatz. The general method we use is a discretisation of the field theory problem into a general Ising model, with the continuous field values being encoded into Ising spin chains. To illustrate the method, and as a simple proof of principle, we have used a quantum annealer to recover the correct profile of various tunnelling solutions. I will discuss this as well as outlining future possibilities. These methods are applicable to many nonperturbative problems.

· Zoom link:  https://zoom.us/j/95375292818?pwd=QS96VkNhbnZDMDJ0d2d6NUJmWVMvdz09