This website requires JavaScript.

Learning Individual Policies in Large Multi-agent Systems through Local Variance Minimization

Tanvi VermaPradeep Varakantham
Dec 2022
摘要
In multi-agent systems with large number of agents, typically thecontribution of each agent to the value of other agents is minimal (e.g.,aggregation systems such as Uber, Deliveroo). In this paper, we consider suchmulti-agent systems where each agent is self-interested and takes a sequence ofdecisions and represent them as a Stochastic Non-atomic Congestion Game (SNCG).We derive key properties for equilibrium solutions in SNCG model withnon-atomic and also nearly non-atomic agents. With those key equilibriumproperties, we provide a novel Multi-Agent Reinforcement Learning (MARL)mechanism that minimizes variance across values of agents in the same state. Todemonstrate the utility of this new mechanism, we provide detailed results on areal-world taxi dataset and also a generic simulator for aggregation systems.We show that our approach reduces the variance in revenues earned by taxidrivers, while still providing higher joint revenues than leading approaches.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答