Fast point cloud generation with diffusion models in high energy physics
- National Energy Research Scientific Computing Center, Berkeley, CA (United States)
- Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States)
- Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
Many particle physics datasets like those generated at colliders are described by continuous coordinates (in contrast to grid points like in an image), respect a number of symmetries (like permutation invariance), and have a stochastic dimensionality. For this reason, standard deep generative models that produce images or at least a fixed set of features are limiting. We introduce a new neural network simulation based on a diffusion model that addresses these limitations named fast point cloud diffusion. We show that our approach can reproduce the complex properties of hadronic jets from proton-proton collisions with competitive precision to other recently proposed models. Additionally, we use a procedure called progressive distillation to accelerate the generation time of our method, which is typically a significant challenge for diffusion models despite their state-of-the-art precision.
- Research Organization:
- Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC), High Energy Physics (HEP); National Energy Research Scientific Computing (NERSC); USDOE
- Grant/Contract Number:
- AC02-05CH11231; HEP-ERCAP0021099
- OSTI ID:
- 1997043
- Alternate ID(s):
- OSTI ID: 2323436
- Journal Information:
- Physical Review. D., Vol. 108, Issue 3; ISSN 2470-0010
- Publisher:
- American Physical Society (APS)Copyright Statement
- Country of Publication:
- United States
- Language:
- English
Similar Records
Optimal transport for a novel event description at hadron colliders
Comparison of point cloud and image-based models for calorimeter fast simulation