Characterising phase transitions in experimental active matter systems with compression algorithms

Light activated colloidal swimmers have been shown to self-organize into dynamic crystalline structures. For our phoretic swimmers, the separation into a high density clustering phase and a low density gas-like phase happens at a surface area fraction above 7%, with a phase transition characterized by particle number fluctuations. The need for ad-hoc order parameters makes locating and characterizing the nature of such dynamical transitions difficult for this class of systems. We will show how compression algorithms can be used to systematically locate and characterize phase transitions in this experimental active matter system, as well as to study the relaxation of the system back to equilibrium.