The low-dimensional Universe: Uncovering hidden information by probabilistic modelling of general, noisily sampled, abstract manifolds

by Dr Marco Canducci (University of Birmingham)

Thursday 4 Dec 2025, 11:30 → 12:30 Europe/Rome

Sala Jappelli (Osservatorio astronomico di Padova)

Filamentary structures (one-dimensional manifolds) are ubiquitous in astronomical data sets. Be it in particle simulations or observations, filaments are always tracers of gravitational and/or hydrodynamical interactions within the observed (simulated) system. However, the recovery of such structures is often complicated by the presence of a large amount of contamination and transverse noise in the observation space. To further complicate the scenario, one-dimensional manifolds (filaments/streams) are generally non-linear and their geometry difficult to extract and model. Thus, in order to study hidden manifolds within the dataset, particular care has to be devoted to contamination removal and transverse noise modelling, while still maintaining accuracy in the recovery of their geometrical structure. 1-DREAM is a toolbox composed of five, new Machine Learning methodologies whose aim is to facilitate manifold extraction in such situations. Each methodology has been designed to address particular issues when dealing with complicated low-dimensional structures convoluted with noise. I will demonstrate its workings on various astronomical cases, ranging from observed Globular Cluster streams, simulated Jellyfish galaxies and Cosmic Web simulations and observations. 

 

Join Zoom Meeting
https://unipd.zoom.us/j/82482883288?pwd=MVvmMWSGaiVrGMyAFOdyQG61OAcI7D.1

Meeting ID: 824 8288 3288
Passcode: 072472