Triumf

Published purpose

The project team will utilize a quantum annealing processors for replacing the computationally expensive first-principles simulation with synthetic data generated by a quantum-assisted Machine Learning model – the Quantum Variational Autoencoder (QVAE). Simulated data enables the analysis of the real data at the Large Hadron Collider (LHC). The LHC, and the experiments that capture and analyze proton-proton collisions produced by the LHC, ATLAS and CMS, are a global scientific effort that resulted in the discovery of the Higgs Boson and the awarding of the 2013 Nobel prize in Physics. This quest continues now to study the Higgs Boson and search for new discoveries, but the cost of generating simulated data threatens the physics reach of upcoming upgrades to the LHC. The QVAEs combine deep encoder and decoder neural networks with a Quantum Boltzmann Machine implementing the latent space – a hidden representation of the data. This latent space can be efficiently sampled utilizing Quantum Annealing devices, creating plentiful and fast simulations of particles and their interactions with the detector.