Modeling Conditional Dependencies in Multiagent Trajectories

Research output: Journal contributionsConference article in journalResearchpeer-review


We study modeling joint densities over sets of random variables (next-step movements of multiple agents) which are conditioned on aligned observations (past trajectories). For this setting, we propose an autoregressive approach to model intra-timestep dependencies, where distributions over joint movements are represented by autoregressive factorizations. In our approach, factors are randomly ordered and estimated with a graph neural network to account for permutation equivariance, while a recurrent neural network encodes past trajectories. We further propose a conditional two-stream attention mechanism, to allow for efficient training of random factorizations. We experiment on trajectory data from professional soccer matches and find that we model low frequency trajectories better than variational approaches.
Original languageEnglish
JournalProceedings of Machine Learning Research
Pages (from-to)10518-10533
Number of pages16
Publication statusPublished - 2022
EventThe 25th International Conference on Artificial Intelligence and Statistics - AISTATS 2022 - Virtual Conference, Online
Duration: 28.03.202230.03.2022
Conference number: 25

Bibliographical note

Funding Information:
We would like to thank Marius Lehne at SAP SE for his essential support and help while writing this paper as well as all reviewers for their valuable feedback.

Publisher Copyright:
Copyright © 2022 by the author(s)