Pietro DeLellis 1,2 , Giovanni Polverino 2 , Gozde Ustuner 2 , Nicole Abaid 3 , Simone Macrı` 4 , Erik M. Bollt 5 & Maurizio Porfiri 2
1 Department of Electrical Engineering and Information Technology, University of Naples Federico II, Naples 80125, Italy,
2 Department of Mechanical and Aerospace Engineering, Polytechnic School of Engineering, New York University, Brooklyn, New York 11201, USA,
3 Department of Engineering Science and Mechanics, Virginia Polytechnic Institute and State University,
Blacksburg, Virginia 24061, USA,
4 Section of Behavioural Neuroscience, Department of Cell Biology and Neuroscience, Istituto
Superiore di Sanita`, Roma 00161, Italy,
5 Department of Mathematics, Clarkson University, Potsdam, New York 13699, USA.
We posit a new geometric perspective to define, detect, and classify inherent patterns of collective behaviour
across a variety of animal species. We show that machine learning techniques, and specifically the isometric
mapping algorithm, allow the identification and interpretation of different types of collective behaviour in
five social animal species. These results offer a first glimpse at the transformative potential of machine
learning for ethology, similar to its impact on robotics, where it enabled robots to recognize objects and
navigate the environment.
2014年1月16日，SCIENTIFIC REPORTS | 4 : 3723 | DOI: 10.1038/srep03723