DeepTreeGANv2: Iterative Pooling of Point Clouds
This work addresses the need for faster generative models in High Energy Physics simulations, but it appears incremental as an extension of an existing method.
The authors tackled the problem of generating large point clouds for particle showers in High Energy Physics to bypass time-consuming simulations, and they extended DeepTreeGAN with a critic to iteratively aggregate point clouds in a tree-based manner, showing it can reproduce complex distributions on the JetNet 150 dataset.
In High Energy Physics, detailed and time-consuming simulations are used for particle interactions with detectors. To bypass these simulations with a generative model, the generation of large point clouds in a short time is required, while the complex dependencies between the particles must be correctly modelled. Particle showers are inherently tree-based processes, as each particle is produced by the decay or detector interaction of a particle of the previous generation. In this work, we present a significant extension to DeepTreeGAN, featuring a critic, that is able to aggregate such point clouds iteratively in a tree-based manner. We show that this model can reproduce complex distributions, and we evaluate its performance on the public JetNet 150 dataset.