Abstract

Contributed Talk - Splinter EScience

Thursday, 12 September 2024, 14:50   (S13)

Spherinator + HiPSter: beyond the 'known unknowns' towards the 'unknown unknowns'

Sebastian Trujillo Gomez, Kai Polsterer, Bernd Doser
Heidelberg Institute for Theoretical Studies (HITS)

Current applications of machine learning in astrophysics focus on teaching machines to perform tasks that require domain experts accurately and extremely efficiently on very large datasets. Although essential in the big data era, this approach is limited by our own intuitions and expectations, and provides at most only answers to the ‘known unknowns’. We propose a new conceptual framework and software tools to assist astronomers in delivering the next scientific breakthroughs by letting the machine learn unbiased representations of complex data ranging from large observational surveys to cosmological simulations. These tools automatically learn low-dimensional representations of complex objects such as galaxies in multimodal data (e.g. images, spectra, datacubes, simulated point clouds, etc.), and provide interactive explorative access to arbitrarily large datasets using a simple graphical interface. Our framework is designed to be interpretable, work seamlessly across datasets regardless of their origin, and provide a path towards discovering the ‘unknown unknowns’.