Quantum Variational Autoencoders for Particle Detector Simulation at the Large Hadron Collider

Physicists all over the world, including those throughout Canada, are gearing up to upgrade the Large Hadron Collider to collide particles more quickly than ever before. The huge datasets collected by enormous detectors like the ATLAS experiment will provide a trove of information about mysterious particles like the Higgs bosons, and might even help researchers find hints of new particles like Dark Matter. But to utilize this enormous dataset, physicists also need to generate simulated data faster than ever before– and traditional computational technologies won’t be able to keep up. Our project proposes to utilize quantum-assisted Machine Learning to demonstrate a solution to this challenge. In our work, we create a powerful machine learning model that can be encoded onto D-Wave quantum computing units, and we can use these devices to generate simulation faster than any other computing technique. A huge amount of work remains to bring this proof of concept to fruition, but we hope our project will bring us a step closer to utilizing quantum computing to make substantial impacts in the realm of scientific computing.

Faculty Supervisor:

Max Swiatlowski

Student:

Partner:

Perimeter Institute

Discipline:

Physics

Sector:

Agriculture; Professional, scientific and technical services

University:

TRIUMF INC.

Program: