Adaptive Software-Defined Network Control Using Kernel-Based Reinforcement Learning: An Empirical Study
Nurakhov Y. Kyzyrkanov A. Otarbay Z. Lebedev D.
December 2025Multidisciplinary Digital Publishing Institute (MDPI)
Applied Sciences (Switzerland)
2025#15Issue 23
Software-defined networking (SDN) requires adaptive control strategies to handle dynamic traffic conditions and heterogeneous network environments. Reinforcement learning (RL) has emerged as a promising solution, yet deep RL methods often face instability, non-stationarity, and reproducibility challenges that limit practical deployment. To address these issues, a kernel-based RL framework is introduced, embedding transition dynamics into reproducing kernel Hilbert spaces (RKHS) and combining kernel ridge regression with policy iteration. This approach enables stable value estimation, enhanced sample efficiency, and interpretability, making it suitable for large-scale and evolving SDN scenarios. Experimental evaluation demonstrates consistent convergence and robustness under traffic variability, with cumulative rewards exceeding those of baseline deep RL methods by more than 22%. The findings highlight the potential of kernel-embedded RL as a practical and theoretically grounded solution for adaptive SDN management and contribute to the broader development of intelligent systems in complex environments.
adaptive network control , intelligent systems , kernel methods , non-stationary environments , policy iteration , reinforcement learning , reproducing kernel Hilbert space , software-defined networking
Text of the article Перейти на текст статьи
Department of Computer Sciences, Faculty of Information Technologies, Al Farabi Kazakh National University, Almaty, 050040, Kazakhstan
Department of Science and Innovation, Astana IT University, Astana, 010000, Kazakhstan
Department of Computer Science, School of Engineering and Digital Sciences, Nazarbayev University, Astana, 010000, Kazakhstan
Department of Computer Sciences
Department of Science and Innovation
Department of Computer Science
10 лет помогаем публиковать статьи Международный издатель
Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026