Modeling Conditional Dependencies in Multiagent Trajectories

Research output: Journal contributionsConference article in journalResearchpeer-review

Authors

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
Volume151
Pages (from-to)10518-10533
Number of pages16
ISSN2640-3498
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
http://aistats.org/aistats2022/

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)

Recently viewed

Researchers

  1. Kerstin Fedder

Publications

  1. Enabling Road Condition Monitoring with an on-board Vehicle Sensor Setup
  2. Fixed-term Contracts and Wages Revisited Using Linked Employer-Employee Data from Germany
  3. Stability analysis of a linear model predictive control and its application in a water recovery process
  4. Probabilistic approach to modelling of recession curves
  5. Some model properties to control a permanent magnet machine using a controlled invariant subspace
  6. Gain Scheduling Controller for Improving Level Control Performance
  7. Understanding the socio-technical aspects of low-code adoption for software development
  8. Scholarly Question Answering Using Large Language Models in the NFDI4DataScience Gateway
  9. Study on the effects of tool design and process parameters on the robustness of deep drawing
  10. Neural correlates of the enactment effect in the brain
  11. A Structure and Content Prompt-based Method for Knowledge Graph Question Answering over Scholarly Data
  12. Children's use of spatial skills in solving two map-reading tasks in real space.
  13. Outperformed by a Computer? - Comparing Human Decisions to Reinforcement Learning Agents, Assigning Lot Sizes in a Learning Factory
  14. Modeling Grounding Processes in Chat-based CSCL
  15. Foreign bias in institutional portfolio allocation
  16. Cross-case knowledge transfer in transformative research: enabling learning in and across sustainability-oriented labs through case reporting
  17. Masked Autoencoder Pretraining for Event Classification in Elite Soccer
  18. Comparison of EKF and TSO for Health Monitoring of a Textile-Based Heater Structure and its Control
  19. Determinants and Outcomes of Dual Distribution:
  20. The identification of up-And downstream industries using input-output tables and a firm-level application to minority shareholdings
  21. Sliding mode and model predictive control for inverse pendulum
  22. A Hermeneutic Interpretation of Concepts in a Cooperative Multicultural Working Project
  23. Multilayer neural networks
  24. Teachers’ temporary support and worked-out examples as elements of scaffolding in mathematical modeling
  25. Self-supervised Siamese Autoencoders
  26. Explaining Disagreement on Interest Rates in a Taylor-Rule Setting
  27. Audio-Hacks
  28. Geodesign as a boundary management process
  29. Octanol-Water Partition Coefficient Measurement by a Simple 1H NMR Method
  30. A New Approach for Optimal Solving Cyclic and Non-Cyclic Bus Drvier Rostering Problems
  31. Online-scheduling using past and real-time data
  32. Semiparametric one-step estimation of a sample selection model with endogenous covariates