多智能体系统中作为延迟可观测性的透明度
Transparency as Delayed Observability in Multi-Agent Systems
AI总结:
本文将多智能体系统中的透明度定义为智能体对环境状态可观测性的延迟,通过多智能体强化学习模拟发现,受限与不受限智能体结果随延迟呈相反趋势,存在使社会福利最大的最优部分透明度机制。
摘要:
在交通网络和股票市场等复杂系统中,透明度是否总是有益的?在多智能体系统中如何定义透明度,以及在何种最优透明度下社会福利达到最高?我们采用基于智能体的视角,将透明度(或其缺失)定义为智能体对环境状态可观测性的延迟,并利用模拟分析延迟对社会福利的影响。为了建模智能体策略在不同延迟下的适应,我们将智能体建模为在模拟环境中在不同延迟下最大化相同目标的学习者。聚焦于受限和不受限两种智能体类型,我们使用多智能体强化学习来评估延迟对智能体结果和社会福利的影响。在模拟金融市场中对我们的框架进行实证演示表明,受限和不受限智能体的结果随延迟呈现相反趋势,并存在一个最优部分透明度机制,在该机制下社会福利达到最大。
英文摘要:
Is transparency always beneficial in complex systems such as traffic networks and stock markets? How is transparency defined in multi-agent systems, and what is its optimal degree at which social welfare is highest? We take an agent-based view to define transparency (or its lacking) as delay in agent observability of environment states, and utilize simulations to analyze the impact of delay on social welfare. To model the adaptation of agent strategies with varying delays, we model agents as learners maximizing the same objectives under different delays in a simulated environment. Focusing on two agent types - constrained and unconstrained, we use multi-agent reinforcement learning to evaluate the impact of delay on agent outcomes and social welfare. Empirical demonstration of our framework in simulated financial markets shows opposing trends in outcomes of the constrained and unconstrained agents with delay, with an optimal partial transparency regime at which social welfare is maximal.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。