Improving Social Welfare While Preserving Autonomy via a Pareto Mediator
This addresses the challenge of balancing social welfare and autonomy in multi-agent systems, offering a novel solution for domains like recommendation games and social dilemmas.
The paper tackles the problem of mediators making decisions for agents with conflicting interests by introducing a Pareto Mediator that improves outcomes for delegating agents without harming them, showing in experiments that it greatly increases social welfare and degrades gracefully to pre-intervention levels when based on incorrect utility models.
Machine learning algorithms often make decisions on behalf of agents with varied and sometimes conflicting interests. In domains where agents can choose to take their own action or delegate their action to a central mediator, an open question is how mediators should take actions on behalf of delegating agents. The main existing approach uses delegating agents to punish non-delegating agents in an attempt to get all agents to delegate, which tends to be costly for all. We introduce a Pareto Mediator which aims to improve outcomes for delegating agents without making any of them worse off. Our experiments in random normal form games, a restaurant recommendation game, and a reinforcement learning sequential social dilemma show that the Pareto Mediator greatly increases social welfare. Also, even when the Pareto Mediator is based on an incorrect model of agent utility, performance gracefully degrades to the pre-intervention level, due to the individual autonomy preserved by the voluntary mediator.