Speaker
Description
In High Energy Physics, detailed and time-consuming simulations are used for particle interactions with detectors. To bypass these simulations with a generative model, it needs to be able to generate large point clouds in a short time while correctly modeling complex dependencies between the particles.
For non-sparse problems on a regular grid, such a model would usually use (De-)Convolution layers to up/down-scale the number of voxels.
In this work, we present novel methods to up/down-scale point clouds. For the up-scaling, we propose the use of a feed-forward network to project each point to multiple. For the down-scaling, we propose a Message Passing Layer that connects a variable number of input points to a fixed number of trainable points.
These operations allow us to construct a Graph GAN that is able to generate such point clouds in a tree-based manner. Particle showers are inherently tree-based processes, as each particle is produced by decays or detector interaction of a particle of the previous generation. We demonstrate the model's performance on the public JetNet and CaloChallange datasets.
Brainstorming idea [abstract]
Up- and down-scaling are critical for generative Neural Networks. While there are established methods for image generation, the generation of point clouds lacks such methods. Which approaches could be worth investigating?
Brainstorming idea [title] | Up/downscaling of Point Clouds |
---|