发表机构
Oxford Robotics Institute; University of Oxford（牛津机器人研究所; 牛津大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出图动力学模型（GDM），用于随机和部分可观测环境中演化拓扑的图世界模型，并引入图分布距离（GDD）度量，实验证明其优于基线并具备大图零样本泛化能力。
AI 中文摘要
基于图的世界模型最近作为一种在关系状态表示上学习转移的方法出现。然而，现有方法大多局限于固定拓扑图或确定性、完全可观测的环境。我们提出了图动力学模型（GDM），一种用于图结构观测的世界模型，旨在处理随机和部分可观测环境中演化拓扑的更一般设置。GDM使用稀疏循环邻接矩阵来建模拓扑更新并执行消息传递，同时采用循环状态空间架构来建模随机转移。此外，我们识别了基于图的世界模型评估中的一个空白，因为现有方法无法提供比较预测和真实联合图状态分布的手段，该联合图状态包含相互依赖的拓扑、节点特征和图特征。因此，我们引入了图分布距离（GDD）度量，它使用带有图核的最大均值差异来全面比较联合下一状态分布。我们在多个环境中评估GDM，包括随机和部分可观测设置。我们证明GDM优于基线模型，并在大图上展现出零样本泛化能力。
英文摘要
Graph-based world models have recently emerged as a means of learning transitions over relational state representations. However, existing approaches are largely limited to fixed-topology graphs or deterministic, fully observable environments. We propose the Graph Dynamics Model (GDM), a world model for graph-structured observations that is designed to handle the more general setting of evolving topologies in stochastic and partially observable environments. The GDM uses a sparse recurrent adjacency matrix to model topology updates and perform message passing, together with a recurrent state-space architecture for modelling stochastic transitions. Furthermore, we identify a gap in the evaluation of graph-based world models, as existing methods do not provide a means of comparing predicted and true distributions over the joint graph state comprising the interdependent topology, node features, and graph features. We therefore introduce the Graph Distribution Distance (GDD) metric, which uses maximum mean discrepancy with a graph kernel to comprehensively compare joint next-state distributions. We evaluate the GDM across several environments, including stochastic and partially observable settings. We demonstrate that GDM outperforms baseline models and displays zero-shot generalisation on large graphs.
