A Decentralized Partially Observable Team Decision Methodology with Delayed Information Sharing
Abstract
We study decentralized partially observable team decision problems with low-rank latent dynamics and unknown system models. The proposed framework combines team-theoretic equivalence with low-rank model representations to address cooperative decision-making in partially observable Markov decision processes without prior knowledge of the transition model. Each team member makes decisions based on local private information and delayed common information shared across the team. Using only this available information, each member learns an approximate low-rank Markov decision process and applies least-squares value iteration to compute its policy. This yields a fully decentralized learning and planning algorithm that requires neither a centralized coordinator nor centralized training. We show that the resulting member-side solutions approximate the centralized team solution: despite partial observability, unknown dynamics, and delayed common information, each member recovers the corresponding component of an approximate team-optimal policy. We further establish finite-sample performance guarantees and derive a corresponding sample-complexity bound for the proposed algorithm.
Related
- Minimum Rate For Partially Observable Linear System with Side Information: LQG Plant and Gaussian-Markov Source
- Ramp Metering Control via Hybrid State Deep Reinforcement Learning in Partially Observable Connected Vehicle Environments
- From Relaxed Indexability to Exact Indexability: A t-Step Approach for Partially Observable Restless Bandits
- Decentralized Decision-Making among Heterogeneous Autonomous Vehicles: An α-Potential Game Framework
- Provably Safe Decentralized Contingency MPC under State-Only Information and Limited Sensing for Nonlinear Multi-agent Systems
BibTeX
@misc{ren2026decentralized,
title = {{A Decentralized Partially Observable Team Decision Methodology with Delayed Information Sharing}},
author = {Xiaoxing Ren and Thomas Parisini and Andreas A. Malikopoulos},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.26783},
eprint = {2609.26783},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.26783},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - A Decentralized Partially Observable Team Decision Methodology with Delayed Information Sharing
AU - Xiaoxing Ren
AU - Thomas Parisini
AU - Andreas A. Malikopoulos
PY - 2026
DA - 2026/09/22
PB - arXiv
DO - 10.48550/arxiv.2609.26783
UR - https://arxiv.org/abs/2609.26783
AB - We study decentralized partially observable team decision problems with low-rank latent dynamics and unknown system models. The proposed framework combines team-theoretic equivalence with low-rank model representations to address cooperative decision-making in partially observable Markov decision processes without prior knowledge of the transition model. Each team member makes decisions based on local private information and delayed common information shared across the team. Using only this available information, each member learns an approximate low-rank Markov decision process and applies least-squares value iteration to compute its policy. This yields a fully decentralized learning and planning algorithm that requires neither a centralized coordinator nor centralized training. We show that the resulting member-side solutions approximate the centralized team solution: despite partial observability, unknown dynamics, and delayed common information, each member recovers the corresponding component of an approximate team-optimal policy. We further establish finite-sample performance guarantees and derive a corresponding sample-complexity bound for the proposed algorithm.
ER -