An Exploration of Learnt Representations of W Jets
arXiv:2109.10919
Abstract
I present a Variational Autoencoder (VAE) trained on collider physics data (specifically boosted jets), with reconstruction error given by an approximation to the Earth Movers Distance (EMD) between input and output jets. This VAE learns a concrete representation of the data manifold, with semantically meaningful and interpretable latent space directions which are hierarchically organized in terms of their relation to physical EMD scales in the underlying physical generative process. The variation of the latent space structure with a resolution hyperparameter provides insight into scale dependent structure of the dataset and its information complexity. I introduce two measures of the dimensionality of the learnt representation that are calculated from this scaling.
arXiv commentPublished version, to appear in ICLR workshop Deep Generative Models for Highly Structured Data. Additional appendices
References in corpus (10)
- Auto-Encoding Variational Bayes
- An Introduction to PYTHIA 8.2
- DELPHES 3, A modular framework for fast simulation of a generic collider experiment
- The Metric Space of Collider Events
- Differential Properties of Sinkhorn Approximation for Learning with Wasserstein Distance
- Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distances
- Exploring the Space of Jets with CMS Open Data
- Linearized Optimal Transport for Collider Events
- Solution of the optimal assignment problem by diagonal scaling algorithms
- Variational Autoencoders for Jet Simulation