arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.08389cs.LG
收藏

高阶拓扑方向性与有向单纯形神经网络

Higher-Order Topological Directionality and Directed Simplicial Neural Networks

Istituto Italiano di Tecnologia(意大利技术研究院);Delft University of Technology(代尔夫特理工大学);Harvard University(哈佛大学)

Manuel Lecha, Andrea Cavallo, Francesca Dominici, Elvin Isufi, Claudio Battiloro

AI总结:

针对高阶拓扑模型强制对称化关系的问题,提出高阶方向性概念并构建 Dir-SNN,以建模有向非单纯形交互,理论与实验证明其表达力和任务性能优势。

摘要:

拓扑深度学习(TDL)已经成为一种范式,用于处理定义在高阶组合拓扑空间(如单纯复形或胞腔复形)上的信号并从中学习。尽管许多复杂系统具有非对称的关系结构,但大多数 TDL 模型都会强制将这些关系对称化。本文首先提出一种新的高阶方向性概念,随后基于该概念设计有向单纯形神经网络(Dir-SNN)。Dir-SNN 是在有向单纯复形上运行的消息传递网络,能够利用单纯形之间有向且可能非对称的交互。据我们所知,这是首个使用高阶方向性概念的 TDL 模型。我们从理论和实验上证明,在区分同构有向图方面,Dir-SNN 比其有向图对应模型具有更强的表达能力。在合成源定位任务上的实验表明,当底层复形为有向时,Dir-SNN 优于无向 SNN;而当底层复形为无向时,二者表现相当。

英文摘要:

Topological Deep Learning (TDL) has emerged as a paradigm to process and learn from signals defined on higher-order combinatorial topological spaces, such as simplicial or cell complexes. Although many complex systems have an asymmetric relational structure, most TDL models forcibly symmetrize these relationships. In this paper, we first introduce a novel notion of higher-order directionality and we then design Directed Simplicial Neural Networks (Dir-SNNs) based on it. Dir-SNNs are message-passing networks operating on directed simplicial complexes able to leverage directed and possibly asymmetric interactions among the simplices. To our knowledge, this is the first TDL model using a notion of higher-order directionality. We theoretically and empirically prove that Dir-SNNs are more expressive than their directed graph counterpart in distinguishing isomorphic directed graphs. Experiments on a synthetic source localization task demonstrate that Dir-SNNs outperform undirected SNNs when the underlying complex is directed, and perform comparably when the underlying complex is undirected.

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。

↑