Deep attention networks reveal the rules of collective motion in zebrafish
F. J. H. Heras, F. Romero-Ferrero, R. C. Hinz, G. G. de Polavieja
This paper uses deep attention networks to model zebrafish collective motion, finding that interactions are primarily alignment with repulsion at close distances, and that fish dynamically select neighbors for decision-making.
bioRxiv · Dec 21, 2018Animal Behavior & Cognition