McAvoy Alex, Kates-Harbeck Julian, Chatterjee Krishnendu, Hilbe Christian
Department of Mathematics, University of Pennsylvania, Philadelphia, PA, USA.
Center for Mathematical Biology, University of Pennsylvania, Philadelphia, PA, USA.
PNAS Nexus. 2022 Jul 27;1(4):pgac141. doi: 10.1093/pnasnexus/pgac141. eCollection 2022 Sep.
Across many domains of interaction, both natural and artificial, individuals use past experience to shape future behaviors. The results of such learning processes depend on what individuals wish to maximize. A natural objective is one's own success. However, when two such "selfish" learners interact with each other, the outcome can be detrimental to both, especially when there are conflicts of interest. Here, we explore how a learner can align incentives with a selfish opponent. Moreover, we consider the dynamics that arise when learning rules themselves are subject to evolutionary pressure. By combining extensive simulations and analytical techniques, we demonstrate that selfish learning is unstable in most classical two-player repeated games. If evolution operates on the level of long-run payoffs, selection instead favors learning rules that incorporate social (other-regarding) preferences. To further corroborate these results, we analyze data from a repeated prisoner's dilemma experiment. We find that selfish learning is insufficient to explain human behavior when there is a trade-off between payoff maximization and fairness.
在许多自然和人工的交互领域中，个体利用过去的经验来塑造未来的行为。这种学习过程的结果取决于个体想要最大化的东西。一个自然的目标是自身的成功。然而，当两个这样的“自私”学习者相互作用时，结果可能对双方都不利，尤其是当存在利益冲突时。在这里，我们探讨学习者如何与自私的对手调整激励机制。此外，我们考虑当学习规则本身受到进化压力时所产生的动态变化。通过结合广泛的模拟和分析技术，我们证明在大多数经典的两人重复博弈中，自私学习是不稳定的。如果进化作用于长期收益水平，那么选择反而有利于纳入社会（他人导向）偏好的学习规则。为了进一步证实这些结果，我们分析了来自重复囚徒困境实验的数据。我们发现，当在收益最大化和公平之间进行权衡时，自私学习不足以解释人类行为。
