This approach seeks to modify other agents'

Cooperation between self-interested individuals is a widespread phenomenon in the natural world, but remains elusive in interactions between artificially intelligent agents.

Cooperation between self-interested individuals is a widespread phenomenon in the natural world, but remains elusive in interactions between artificially intelligent agents. An emerging literature on opponent shaping has demonstrated the ability to reach prosocial outcomes by influencing the learning of other agents. However, such methods differentiate through the learning step of other agents or optimize for meta-game dynamics, which rely on privileged access to opponents' learning algorithms or exponential sample complexity, respectively. To provide a learning rule-agnostic and sample-efficient alternative, we introduce Reciprocators, reinforcement learning agents which are intrinsically motivated to reciprocate the influence of opponents' actions on their returns. This approach seeks to modify other agents' Q -values by increasing their return following beneficial actions (with respect to the Reciprocator) and decreasing it after detrimental actions, guiding them towards mutually beneficial actions without directly differentiating through a model of their policy.

Emergent cooperation among self-interested individuals is a widespread phenomenon in the natural world, but remains elusive in interactions between artificially intelligent agents. Instead, na\"ive reinforcement learning algorithms typically converge to Pareto-dominated outcomes in even the simplest of social dilemmas. An emerging class of opponent-shaping methods have demonstrated the ability to reach prosocial outcomes by influencing the learning of other agents. However, they rely on higher-order derivatives through the predicted learning step of other agents or learning meta-game dynamics, which in turn rely on stringent assumptions over opponent learning rules or exponential sample complexity, respectively. To provide a learning rule-agnostic and sample-efficient alternative, we introduce Reciprocators, reinforcement learning agents which are intrinsically motivated to reciprocate the influence of an opponent's actions on their returns. This approach effectively seeks to modify other agents' $Q$-values by increasing their return following beneficial actions (with respect to the Reciprocator) and decreasing it after detrimental actions, guiding them towards mutually beneficial actions without attempting to directly shape policy updates. We show that Reciprocators can be used to promote cooperation in a variety of temporally extended social dilemmas during simultaneous learning.

Aggregated phenomena in social sciences and economi cs are highly dependent on the way individuals interact. To help understanding the interplay betwe en socio-economic activities and underlying social networks, this paper studies a sequential prisoner's dilemma with binary choice. It proposes an analytical and computational insight about the role of endogenous networks in emergence and sustainability of cooperation and exhibits an alternative to the choice and refusal mechanism that is often proposed to explain cooperation. The study fo cuses on heterogeneous equilibriums and emergence of cooperation from an all-defector state that are the two stylized facts that this model successfully reconstructs.