Search for: All records

Abstract Particle collisions at accelerators like the Large Hadron Collider (LHC), recorded by experiments such as ATLAS and CMS, enable precise standard model measurements and searches for new phenomena. Simulating these collisions significantly influences experiment design and analysis but incurs immense computational costs, projected at millions of CPU-years annually during the high luminosity LHC (HL-LHC) phase. Currently, simulating a single event with Geant4 consumes around 1000 CPU seconds, with calorimeter simulations especially demanding. To address this, we propose a conditioned quantum-assisted generative model, integrating a conditioned variational autoencoder (VAE) and a conditioned restricted Boltzmann machine (RBM). Our RBM architecture is tailored for D-Wave’s Pegasus-structured advantage quantum annealer for sampling, leveraging the flux bias for conditioning. This approach combines classical RBMs as universal approximators for discrete distributions with quantum annealing’s speed and scalability. We also introduce an adaptive method for efficiently estimating effective inverse temperature, and validate our framework on Dataset 2 of CaloChallenge. 

As we approach the High Luminosity Large Hadron Collider (HL-LHC) set to begin collisions by the end of this decade, it is clear that the computational demands of traditional collision simulations have become untenably high. Current methods, relying heavily on first-principles Monte Carlo simulations for event showers in calorimeters, are estimated to require millions of CPU-years annually, a demand that far exceeds current capabilities. This bottleneck presents a unique opportunity for breakthroughs in computational physics through the integration of generative AI with quantum computing technologies. We propose a Quantum-Assisted deep generative model. In particular, we combine a variational autoencoder (VAE) with a Restricted Boltzmann Machine (RBM) embedded in its latent space as a prior. The RBM in latent space provides further expressiveness compared to legacy VAE where the prior is a fixed Gaussian distribution. By crafting the RBM couplings, we leverage D-Wave’s Quantum Annealer to significantly speed up the shower sampling time. By combining classical and quantum computing, this framework sets a path towards utilizing large-scale quantum simulations as priors in deep generative models and demonstrate their ability to generate high-quality synthetic data for the HL-LHC experiments.