STEMFold: Stochastic temporal manifold for multi-agent interactions in the presence of hidden agents

Hemant Kumawat, Biswadeep Chakraborty, Saibal Mukhopadhyay
Proceedings of the 6th Annual Learning for Dynamics & Control Conference, PMLR 242:1427-1439, 2024.

Abstract

Learning accurate, data-driven predictive models for multiple interacting agents following unknown dynamics is crucial in many real-world physical and social systems. In many scenarios, dynamics prediction must be performed under incomplete observations, i.e., only a subset of agents are known and observable from a larger topological system while the behaviors of the unobserved agents and their interactions with the observed agents are not known. When only incomplete observations of a dynamical system are available, so that some states remain hidden, it is generally not possible to learn a closed-form model in these variables using either analytic or data-driven techniques. In this work, we propose STEMFold, a spatiotemporal attention-based generative model, to learn a stochastic manifold to predict the underlying unmeasured dynamics of the multi-agent system from observations of only visible agents. Our analytical results motivate STEMFold design using a spatiotemporal graph with time anchors to effectively map the observations of visible agents to a stochastic manifold with no prior information about interaction graph topology. We empirically evaluated our method on two simulations and two real-world datasets, where it outperformed existing networks in predicting complex multiagent interactions, even with many unobserved agents.

Cite this Paper

BibTeX
@InProceedings{pmlr-v242-kumawat24a, title = {{STEMFold}: {S}tochastic temporal manifold for multi-agent interactions in the presence of hidden agents}, author = {Kumawat, Hemant and Chakraborty, Biswadeep and Mukhopadhyay, Saibal}, booktitle = {Proceedings of the 6th Annual Learning for Dynamics & Control Conference}, pages = {1427--1439}, year = {2024}, editor = {Abate, Alessandro and Cannon, Mark and Margellos, Kostas and Papachristodoulou, Antonis}, volume = {242}, series = {Proceedings of Machine Learning Research}, month = {15--17 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v242/kumawat24a/kumawat24a.pdf}, url = {https://proceedings.mlr.press/v242/kumawat24a.html}, abstract = {Learning accurate, data-driven predictive models for multiple interacting agents following unknown dynamics is crucial in many real-world physical and social systems. In many scenarios, dynamics prediction must be performed under incomplete observations, i.e., only a subset of agents are known and observable from a larger topological system while the behaviors of the unobserved agents and their interactions with the observed agents are not known. When only incomplete observations of a dynamical system are available, so that some states remain hidden, it is generally not possible to learn a closed-form model in these variables using either analytic or data-driven techniques. In this work, we propose STEMFold, a spatiotemporal attention-based generative model, to learn a stochastic manifold to predict the underlying unmeasured dynamics of the multi-agent system from observations of only visible agents. Our analytical results motivate STEMFold design using a spatiotemporal graph with time anchors to effectively map the observations of visible agents to a stochastic manifold with no prior information about interaction graph topology. We empirically evaluated our method on two simulations and two real-world datasets, where it outperformed existing networks in predicting complex multiagent interactions, even with many unobserved agents.} }
Endnote
%0 Conference Paper %T STEMFold: Stochastic temporal manifold for multi-agent interactions in the presence of hidden agents %A Hemant Kumawat %A Biswadeep Chakraborty %A Saibal Mukhopadhyay %B Proceedings of the 6th Annual Learning for Dynamics & Control Conference %C Proceedings of Machine Learning Research %D 2024 %E Alessandro Abate %E Mark Cannon %E Kostas Margellos %E Antonis Papachristodoulou %F pmlr-v242-kumawat24a %I PMLR %P 1427--1439 %U https://proceedings.mlr.press/v242/kumawat24a.html %V 242 %X Learning accurate, data-driven predictive models for multiple interacting agents following unknown dynamics is crucial in many real-world physical and social systems. In many scenarios, dynamics prediction must be performed under incomplete observations, i.e., only a subset of agents are known and observable from a larger topological system while the behaviors of the unobserved agents and their interactions with the observed agents are not known. When only incomplete observations of a dynamical system are available, so that some states remain hidden, it is generally not possible to learn a closed-form model in these variables using either analytic or data-driven techniques. In this work, we propose STEMFold, a spatiotemporal attention-based generative model, to learn a stochastic manifold to predict the underlying unmeasured dynamics of the multi-agent system from observations of only visible agents. Our analytical results motivate STEMFold design using a spatiotemporal graph with time anchors to effectively map the observations of visible agents to a stochastic manifold with no prior information about interaction graph topology. We empirically evaluated our method on two simulations and two real-world datasets, where it outperformed existing networks in predicting complex multiagent interactions, even with many unobserved agents.
APA
Kumawat, H., Chakraborty, B. & Mukhopadhyay, S.. (2024). STEMFold: Stochastic temporal manifold for multi-agent interactions in the presence of hidden agents. Proceedings of the 6th Annual Learning for Dynamics & Control Conference, in Proceedings of Machine Learning Research 242:1427-1439 Available from https://proceedings.mlr.press/v242/kumawat24a.html.

Related Material