Improved Structural Discovery and Representation Learning of Multi-Agent Data

Jennifer Hobbs, Matthew Holbrook, Nathan Frank, Long Sha, Patrick Lucey

Keywords: multi-agent, gaussian mixture, permutation learning, representation learning, group structure

TL;DR: We propose an improved approach to discovering the group structure and ordered representation of multi-agent data

Abstract: Central to all machine learning algorithms is data representation. For multi-agent systems, selecting a representation which adequately captures the interactions among agents is challenging due to the latent group structure which tends to vary depending on various contexts. However, in multi-agent systems with strong group structure, we can simultaneously learn this structure and map a set of agents to a consistently ordered representation for further learning. In this paper, we present a dynamic alignment method which provides a robust ordering of structured multi-agent data which allows for representation learning to occur in a fraction of the time of previous methods. We demonstrate the value of this approach using a large amount of soccer tracking data from a professional league.