Social learning spontaneously emerges by searching optimal heuristics with deep reinforcement learning
This work addresses the optimization of social learning strategies for agents in cooperative environments, with potential applications in multi-agent systems and social network analysis, though it is incremental in applying reinforcement learning to this domain.
The study tackled the problem of how social learning strategies evolve and what optimal strategies exist in cooperative games, finding that deep reinforcement learning agents spontaneously learn effective social learning concepts and outperform traditional baseline strategies in mean payoff.
How have individuals of social animals in nature evolved to learn from each other, and what would be the optimal strategy for such learning in a specific environment? Here, we address both problems by employing a deep reinforcement learning model to optimize the social learning strategies (SLSs) of agents in a cooperative game in a multi-dimensional landscape. Throughout the training for maximizing the overall payoff, we find that the agent spontaneously learns various concepts of social learning, such as copying, focusing on frequent and well-performing neighbors, self-comparison, and the importance of balancing between individual and social learning, without any explicit guidance or prior knowledge about the system. The SLS from a fully trained agent outperforms all of the traditional, baseline SLSs in terms of mean payoff. We demonstrate the superior performance of the reinforcement learning agent in various environments, including temporally changing environments and real social networks, which also verifies the adaptability of our framework to different social settings.