Trust-Based Social Learning for Communication (TSLEC) Protocol Evolution in Multi-Agent Reinforcement Learning

TL;DR

This paper presents TSLEC, a trust-based social learning framework that significantly accelerates emergent communication in multi-agent systems by enabling agents to teach successful strategies based on learned trust, resulting in faster convergence and robust protocols.

Contribution

Introduces TSLEC, a novel trust-based social learning framework that improves convergence speed and protocol robustness in multi-agent communication through explicit teaching and trust modeling.

Findings

Trust-based social learning reduces episodes-to-convergence by 23.9%.

Trust scores strongly correlate with teaching quality (r=0.743).

Protocols remain robust under dynamic objectives (Phi > 0.867).

Abstract

Emergent communication in multi-agent systems typically occurs through independent learning, resulting in slow convergence and potentially suboptimal protocols. We introduce TSLEC (Trust-Based Social Learning with Emergent Communication), a framework where agents explicitly teach successful strategies to peers, with knowledge transfer modulated by learned trust relationships. Through experiments with 100 episodes across 30 random seeds, we demonstrate that trust-based social learning reduces episodes-to-convergence by 23.9% (p < 0.001, Cohen's d = 1.98) compared to independent emergence, while producing compositional protocols (C = 0.38) that remain robust under dynamic objectives (Phi > 0.867 decoding accuracy). Trust scores strongly correlate with teaching quality (r = 0.743, p < 0.001), enabling effective knowledge filtering. Our results establish that explicit social learning fundamentally accelerates emergent communication in multi-agent coordination.