Evaluating Generalization in Multiagent Systems using Agent-Interaction Graphs

Evaluating Generalization in Multiagent Systems using Agent-Interaction Graphs
复制标题

使用代理交互图评估多代理系统中的泛化

DOI:
--
复制
发表时间:
2018
期刊:
--
影响因子:
--
通讯作者:
Harrison Edwards
Harrison Edwards
中科院分区:
--
文献类型:
--
作者:
Aditya Grover;Maruan Al;Jayesh K. Gupta;Yuri Burda;Harrison Edwards

文献摘要

参考文献

被引文献

相似文献

从代理之间的交互中学习是多代理系统中推理的关键组成部分。根据下游任务的不同,可以有多个标准来评估学习的泛化性能。在这项工作中,我们提出了一个新的框架,用于评估多智能体系统的泛化基于代理交互图。代理交互图将代理建模为节点,将交互建模为参与代理之间的超边。使用这个抽象的数据结构,我们定义了三个概念的泛化多智能体系统的学习原则评价。
Learning from interactions between agents is a key component for inference in multiagent systems. Depending on the downstream task, there could be multiple criteria for evaluating the generalization performance of learning. In this work, we propose a novel framework for evaluating generalization in multiagent systems based on agent-interaction graphs. An agent-interaction graph models agents as nodes and interactions as hyper-edges between participating agents. Using this abstract data structure, we define three notions of generalization for principled evaluation of learning in multiagent systems.
DOI: --
发表时间: 2018
期刊:
影响因子: --
作者:
Jiexin Wang;Eiji Uchibe;Kenji Doya;Eiji Uchibe;Eiji Uchibe
通讯作者: Eiji Uchibe