The paper introduces a general theoretical framework for fibrations on graphs labeled by a commutative monoid, extending the classic theory of graph fibrations to weighted and algebraically labeled graphs. It also accommodates approximate fibrations and demonstrates how this framework can be used to compress arbitrary neural networks, including CNNs, providing a solid theoretical basis for recent findings on fibration symmetries in geometric deep learning.
By Paolo Boldi
Posted by Ameya Velingker, Research Scientist, Google Research, and Balaji Venkatachalam, Software Engineer, Google Graphs , in which objects and their relations are represented as nodes (or vertices) and edges (or links) between pairs of nodes, are ubiquitous in computing and machine learning (ML). For example, social networks, road networks, and molecular structure and interactions are all domains in which underlying datasets have a natural graph structure.
By Google AI
The paper establishes a precise mathematical link between graph surgery and the do‑operator in deterministic acyclic structural causal models. It shows that deleting arrows in a graph corresponds exactly to replacing the associated mechanisms with constants, proving that “Graph(F^\iota)=Surg(Graph(F),T_\iota)”. The authors further characterize when this equality holds for the full graph, define the intervened model, and demonstrate how sequential interventions combine, concluding that an outcome depends only on interventions at its actual dependency ancestors.
By Satpreet Makhija
The paper studies graphons—limits of graph sequences—by transforming sparse finite graphs into their line graphs. It identifies a property called the square‑degree property that ensures sparse graphs produce dense line graphs, allowing dense‑graph limit results to apply. Examples include star graphs and superlinear preferential attachment graphs, whose line graphs yield non‑zero graphons, while dense graphs such as Erdős–Rényi produce sparse line graphs and the zero graphon.
By Sevvandi Kandanaarachchi, Cheng Soon Ong
arXiv:2607. 01057v1 Announce Type: cross Abstract: We study a broad class of graphical models whose independencies correspond to vertex separation in mixed graphs with directed, undirected, and bidirected edges, that are capable of encoding independence structures arising from feedback, latent and selection mechanisms.
By Christopher Meek, Kayvan Sadeghi
arXiv:2603. 14846v3 Announce Type: replace Abstract: We define an information-complexity property for aggregation functions, capturing a vast range of practical aggregations, and prove that any Message-Passing Graph Neural Network (MP-GNN) model with such aggregations induces only a polynomial number of equivalence classes on all graphs - while the number of non-isomorphic graphs is super-exponential (in number of vertices).
By Eran Rosenbluth
The paper introduces a new (pseudo)-metric called graph homomorphism distortion, which quantifies how much node features must be distorted when mapping one attributed graph to another. This metric builds on metric geometry concepts and leverages graph homomorphisms to capture both structure and features. The authors show that the metric can be efficiently approximated, offers fine-grained graph distinction, and can serve as a structural encoding that enhances graph neural network predictions.
By Martin Carrasco, Olga Zaghen, Kavir Sumaraj, Erik Bekkers, Bastian Rieck
arXiv:2610. 07503v1 Announce Type: cross Abstract: Two populations of graphs often have to be compared without any correspondence between their vertices, for instance when networks come from different communities, or when a graph generative model is evaluated against held-out graphs.
By Soham Dan
In this paper, we study the theoretical computational complexity of counting trees and graphs that are classified positively by a graph neural networks, or said equivalently that satisfy a given formu...
arXiv:2606. 07728v1 Announce Type: new Abstract: It is well established that ReLU networks define continuous piecewise-linear functions, and that their linear regions are polyhedra in the input space.
By Blake B. Gaines, Jinbo Bi
arXiv:2607. 12026v1 Announce Type: cross Abstract: Finite groups are rigid algebraic objects, whose Cayley graphs expose a rich network geometry through which group-theoretic structure can be measured, compared, and learned.
By Rashid Barket, Enrico Grimaldi, Yacoub Hendi, Edward Hirst, Adam Onus, Harmeet Singh
arXiv:2605. 00725v2 Announce Type: replace Abstract: Topological neural networks have emerged as effective tools for modeling higher-order relational structures beyond pairwise graphs, including hypergraphs, simplicial complexes, and cell complexes.
By Jiawen Chen, Qi Shao, Zhiqiang Ge, Duxin Chen, Wenwu Yu