Data-Driven Inverse Reinforcement Learning for Networked 2-D Markov Jump Systems
IEEE Internet of Things Journal · 2026-08 · 13 (16) · 37196-37205
DOI: 10.1109/jiot.2026.3701745 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Internet of Things Journal, 2026-08. Verfasst von Yang Liu und 4 weiteren Autor:innen. Die Arbeit nennt 16 Literaturangaben. Gefördert von: Cross-Integration Innovation Team of Modern Applied Mathematics and Life Sciences in Yunnan Province.
Abstract (Originalsprache)
This article investigates inverse reinforcement learning (IRL) for continuous-time networked 2-D Markov jump systems (2-D MJSs). In many IoT applications, control policies are generated by expert controllers or cloud-based optimization modules, while the underlying performance objectives are implicit and therefore unavailable to distributed edge nodes. To address this challenge, an IRL framework is developed to reconstruct mode-dependent cost functions from observed state and input trajectories. By exploiting the coupled algebraic Riccati equation structure associated with 2-D MJSs, inverse optimality conditions are reformulated across both propagation dimensions and stochastic modes. The proposed approach establishes asymptotic mean-square stability and convergence of the recovered cost matrices, and it reveals fundamental identifiability and homogeneity properties that characterize reward ambiguity in inverse learning. To improve scalability and reduce model dependence, a fully data-driven IRL algorithm is further proposed, enabling cost reconstruction without explicit knowledge of system dynamics. Numerical and physics-grounded simulation examples demonstrate the effectiveness of the proposed methods.
Förderung
- Cross-Integration Innovation Team of Modern Applied Mathematics and Life Sciences in Yunnan Province
Zitieren
APA
Liu, Y., Cheng, J., Yan, M., Zhang, D., & Zhang, B. (2026). Data-Driven Inverse Reinforcement Learning for Networked 2-D Markov Jump Systems. IEEE Internet of Things Journal, 13(16), 37196-37205. https://doi.org/10.1109/jiot.2026.3701745
Vancouver
Liu Y, Cheng J, Yan M, Zhang D, Zhang B. Data-Driven Inverse Reinforcement Learning for Networked 2-D Markov Jump Systems. IEEE Internet of Things Journal. 2026;13(16):37196-37205. doi:10.1109/jiot.2026.3701745
BibTeX
@article{Liu2026,
author = {Liu, Yang and Cheng, Jun and Yan, Mei and Zhang, Dan and Zhang, Bin},
title = {Data-Driven Inverse Reinforcement Learning for Networked 2-D Markov Jump Systems},
journal = {IEEE Internet of Things Journal},
year = {2026},
volume = {13},
number = {16},
pages = {37196--37205},
doi = {10.1109/jiot.2026.3701745}
}
EUVIMED