EUVIMEDEUVIMED entdecken
Crossref

Advancements in Explainable Reinforcement Learning Algorithms

Avinash M. Pawar, Nitin Sherje

International Journal on Advanced Computer Engineering and Communication Technology · 2025

DOI: 10.65521/ijacect.v12i1.128 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal on Advanced Computer Engineering and Communication Technology, 2025. Verfasst von Avinash M. Pawar und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reinforcement Learning (RL) has demonstrated remarkable success in complex decision-making tasks; however, the black-box nature of many RL models limits their interpretability, hindering trust, transparency, and real-world deployment. Explainable Reinforcement Learning (XRL) seeks to bridge this gap by integrating interpretability mechanisms into RL frameworks. This paper reviews recent advancements in XRL, including model-agnostic explainability methods, intrinsically interpretable RL architectures, and human-in-the-loop strategies. We discuss techniques such as policy visualization, reward decomposition, attention mechanisms, and counterfactual explanations, highlighting their effectiveness in providing insights into agent behavior. Additionally, we explore the challenges and future directions in XRL, particularly in balancing explainability with performance and generalizability. As RL continues to be applied in high-stakes domains such as healthcare, finance, and autonomous systems, enhancing its interpretability remains crucial for broader adoption and ethical AI development.

Zitieren

APA

Pawar, A. M., & Sherje, N. (2025). Advancements in Explainable Reinforcement Learning Algorithms. International Journal on Advanced Computer Engineering and Communication Technology. https://doi.org/10.65521/ijacect.v12i1.128

Vancouver

Pawar AM, Sherje N. Advancements in Explainable Reinforcement Learning Algorithms. International Journal on Advanced Computer Engineering and Communication Technology. 2025. doi:10.65521/ijacect.v12i1.128

BibTeX
@article{Pawar2025,
  author = {Pawar, Avinash M. and Sherje, Nitin},
  title = {Advancements in Explainable Reinforcement Learning Algorithms},
  journal = {International Journal on Advanced Computer Engineering and Communication Technology},
  year = {2025},
  doi = {10.65521/ijacect.v12i1.128}
}
Kostenlos starten →