Unsupervised classification reveals new evolutionary pathways

TL;DR

This study uses a large galaxy dataset and unsupervised clustering to identify 12 galaxy classes, revealing potential new evolutionary pathways and challenging existing galaxy classification assumptions.

Contribution

The paper introduces a novel application of unsupervised clustering to high-redshift galaxy data, uncovering new galaxy classes and evolutionary insights.

Findings

Identification of 12 galaxy classes from VIPERS survey data

Discovery of red nuggets at intermediate redshift

Challenging of mid-infrared AGN selection methods

Abstract

While we already seem to have a general scenario of the evolution of different types of galaxies, a complete and satisfactory understanding of the processes that led to the formation of all the variety of today's galaxy types is still beyond our reach. To solve this problem, we need both large datasets reaching high redshifts and novel methodologies for dealing with them. The VIPERS survey statistical power, which observed $\sim90,000$ galaxies at $z > 0.5$, and the application of an unsupervised clustering algorithm allowed us to distinguish 12 galaxy classes. Studies of their environmental dependence indicate that this classification may actually reflect different galaxy evolutionary paths. For instance, a class of the most passive red galaxies gathers galaxies $\sim20\%$ smaller than other red galaxies of a similar stellar mass, revealing the first sample of red nuggets at intermediate redshift. On the other end, a class of blue dwarf galaxies is composed mainly of AGN, challenging commonly used mid-infrared AGN selections.