Multi-expert policy distillation guided proximal policy optimization for efficient cooperation emergence in spatial public goods games

Chanchan Li, Zhaoqilin Yang, Wensheng Jia, Hongxin Zhao and Xin Wang

Chaos, Solitons & Fractals, 2026, vol. 202, issue P1

Abstract: Our work is inspired by the UNESCO-intangible heritage Tribunal de las Aguas de Valencia. Based on this inspiration, an innovative framework, namely the Multi-Expert Policy Distillation guided Proximal Policy Optimization (MEPD-PPO), is proposed. This novel approach resolves fundamental cooperation challenges in spatial public goods games under subcritical conditions. It achieves this by integrating knowledge distillation from diverse experts with deep reinforcement learning. Our framework uniquely employs temperature-scaled KL divergence minimization for efficient policy transfer and the convolution operation for enhanced spatial feature extraction. The proposed MEPD-PPO demonstrates breakthrough capabilities in establishing sustainable cooperation where conventional methods fail. Experimental validation confirms our method’s superior performance in achieving lower cooperation thresholds, faster convergence, and exceptional robustness across adversarial conditions. This research establishes a new paradigm for leadership-guided collective action systems. Multiple experts with diverse knowledge guide and bootstrap the emergence of cooperation in our MEPD-PPO framework. This mechanism bridges evolutionary principles with multi-agent learning, enabling more resilient socio-technical infrastructures.

Keywords: Spatial public goods games; Deep reinforcement learning; Proximal policy optimization; Multi-expert policy distillation; Convolution operation (search for similar items in EconPapers)
Date: 2026
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0960077925014900
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:chsofr:v:202:y:2026:i:p1:s0960077925014900

DOI: 10.1016/j.chaos.2025.117477

Access Statistics for this article

Chaos, Solitons & Fractals is currently edited by Stefano Boccaletti and Stelios Bekiros

More articles in Chaos, Solitons & Fractals from Elsevier
Bibliographic data for series maintained by Thayer, Thomas R. ().