Learning to control active matter

The study of active matter has revealed novel non-equilibrium collective behaviors, illustrating their potential as a new materials platform. However, most work treat unregulated systems with uniform microscopic energy input, which we refer to activity. In contrast, functionality in biological results from regulating and controlling activity locally over space time, only recently become experimentally possible for engineered matter. Designing requires navigation the high-dimensional spatio-temporal patterns, but brute force approaches are unlikely be successful without system-specific intuition. Here, apply reinforcement learning task inducing net transport specific direction simulated system Vicsek-like self-propelled disks using spotlight that increases locally. resulting time-varying patterns learned exploit distinct physics strong weak coupling regimes. Our shows how can reveal physically interpretable protocols behavior systems.