EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Model-agnostic solutions for deep reinforcement learning in non-ergodic contexts

Bert Verbruggen, Arne Vanhoyweghen, Vincent Ginis

Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · 19.11.2026 · 384 (2334)

DOI: 10.1098/rsta.2025.0154 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, online erschienen am 19.11.2026. Verfasst von Bert Verbruggen und 2 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Research Foundation Flanders.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reinforcement learning (RL) remains a central optimization framework in machine learning. Although RL agents can converge to optimal solutions, the definition of ‘optimality’ depends on the environment's statistical properties. The Bellman equation, central to most RL algorithms, is formulated in terms of expected values of future rewards. However, when ergodicity is broken, long-term outcomes depend on the specific trajectory rather than on the ensemble average. In such settings, the ensemble average diverges from the time-average growth experienced by individual agents, with expected-value formulations yielding systematically suboptimal policies. Prior studies demonstrated that traditional RL architectures fail to recover the true optimum in non-ergodic environments. We extend this analysis to deep RL implementations and show that these, too, produce suboptimal policies under non-ergodic dynamics. We show that this limitation can be addressed by treating non-ergodic decision problems as finite trajectory-exposure problems rather than as isolated one-step optimization tasks. By exposing the agent to compounded trajectories during training, the learned value estimates become more consistent with the process's intrinsic time growth. This improvement does not require altering the environmental feedback, such as reward transformations or modified objective functions, but arises naturally from the agent's exposure to temporal trajectories. Our results contribute to the growing body of research on RL methods for non-ergodic systems. This article is part of the theme issue ‘Ergodicity and ergodicity breaking in the sciences’.

Förderung

  • Research Foundation Flanders

Zitieren

APA

Verbruggen, B., Vanhoyweghen, A., & Ginis, V. (2026). Model-agnostic solutions for deep reinforcement learning in non-ergodic contexts. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 384(2334). https://doi.org/10.1098/rsta.2025.0154

Vancouver

Verbruggen B, Vanhoyweghen A, Ginis V. Model-agnostic solutions for deep reinforcement learning in non-ergodic contexts. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences. 2026;384(2334). doi:10.1098/rsta.2025.0154

BibTeX
@article{Verbruggen2026,
  author = {Verbruggen, Bert and Vanhoyweghen, Arne and Ginis, Vincent},
  title = {Model-agnostic solutions for deep reinforcement learning in non-ergodic contexts},
  journal = {Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences},
  year = {2026},
  volume = {384},
  number = {2334},
  doi = {10.1098/rsta.2025.0154}
}
Kostenlos starten →