Emergent Cooperation under Uncertain Incentive Alignment
This addresses the challenge of developing cooperative AI in real-world settings with sparse interactions and uncertain incentives, representing an incremental improvement by applying known mechanisms to a specific scenario.
The paper tackled the problem of cooperation among reinforcement learning agents under uncertain incentive alignment and infrequent encounters, finding that uncertainty lowers cooperative behavior, but reputation mechanisms and intrinsic rewards boost near-optimal cooperation in cooperative and mixed-motive environments.
Understanding the emergence of cooperation in systems of computational agents is crucial for the development of effective cooperative AI. Interaction among individuals in real-world settings are often sparse and occur within a broad spectrum of incentives, which often are only partially known. In this work, we explore how cooperation can arise among reinforcement learning agents in scenarios characterised by infrequent encounters, and where agents face uncertainty about the alignment of their incentives with those of others. To do so, we train the agents under a wide spectrum of environments ranging from fully competitive, to fully cooperative, to mixed-motives. Under this type of uncertainty we study the effects of mechanisms, such as reputation and intrinsic rewards, that have been proposed in the literature to foster cooperation in mixed-motives environments. Our findings show that uncertainty substantially lowers the agents' ability to engage in cooperative behaviour, when that would be the best course of action. In this scenario, the use of effective reputation mechanisms and intrinsic rewards boosts the agents' capability to act nearly-optimally in cooperative environments, while greatly enhancing cooperation in mixed-motive environments as well.