CellCLAT: Preserving Topology and Trimming Redundancy in Self-Supervised Cellular Contrastive Learning

2025/05/27 by Bin Qin, Qirui Ji, Qin, Bin +10 · 1 citation
Computer Science · #Advanced Graph Neural Networks #Artificial Intelligence (cs.AI) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Topological and Geometric Data Analysis

paper · pdf · doi:10.48550/arxiv.2505.21587

openalex publication_date 2025/05/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Self-supervised topological deep learning (TDL) represents a nascent but underexplored area with significant potential for modeling higher-order interactions in simplicial complexes and cellular complexes to derive representations of unlabeled graphs. Compared to simplicial complexes, cellular complexes exhibit greater expressive power. However, the advancement in self-supervised learning for cellular TDL is largely hindered by two core challenges: extrinsic structural constraints inherent to cellular complexes, and intrinsic semantic redundancy in cellular representations. The first challenge highlights that traditional graph augmentation techniques may compromise the integrity of higher-order cellular interactions, while the second underscores that topological redundancy in cellular complexes potentially diminish task-relevant information. To address these issues, we introduce Cellular Complex Contrastive Learning with Adaptive Trimming (CellCLAT), a twofold framework designed to adhere to the combinatorial constraints of cellular complexes while mitigating informational redundancy. Specifically, we propose a parameter perturbation-based augmentation method that injects controlled noise into cellular interactions without altering the underlying cellular structures, thereby preserving cellular topology during contrastive learning. Additionally, a cellular trimming scheduler is employed to mask gradient contributions from task-irrelevant cells through a bi-level meta-learning approach, effectively removing redundant topological elements while maintaining critical higher-order semantics. We provide theoretical justification and empirical validation to demonstrate that CellCLAT achieves substantial improvements over existing self-supervised graph learning methods, marking a significant attempt in this domain.

Citations

• TopoTune : A Framework for Generalized Combinatorial Complex Neural Networks
• CliquePH: Higher-Order Information for Graph Neural Networks through Persistent Homology on Clique Graphs
• Topological Blindspots: Understanding and Extending Topological Deep Learning Through the Lens of Expressivity
• Hierarchical Topology Isomorphism Expertise Embedded Graph Contrastive Learning
• Rethinking Dimensional Rationale in Graph Contrastive Learning from Causal Perspective
• On the Expressivity of Persistent Homology in Graph Learning
• Topological Deep Learning: Going Beyond Graph Data
• Topological Relational Learning on Graphs
• Weisfeiler and Lehman Go Cellular: CW Networks
• Graph Contrastive Learning Automated
• Adversarial Graph Augmentation to Improve Graph Contrastive Learning
• Weisfeiler and Lehman Go Topological: Message Passing Simplicial\n Networks
• Principled Simplicial Neural Networks for Trajectory Prediction
• Graph Contrastive Learning with Augmentations
• Paper-author bipartite graph from Semantic Scholar
• Cell Complex Neural Networks
• TUDataset: A collection of benchmark datasets for learning with graphs
• Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere
• InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization
• Graph Neural Networks for Social Recommendation
• Self-Supervised Generalisation with Meta Auxiliary Learning
• DARTS: Differentiable Architecture Search
• Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination
• graph2vec: Learning Distributed Representations of Graphs
• Neural Message Passing for Quantum Chemistry
• Categorical Reparameterization with Gumbel-Softmax
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• node2vec: Scalable Feature Learning for Networks
• Amortized O(|V|)-Delay Algorithm for Listing Chordless Cycles in Undirected Graphs

Cited by

• HTG-GCL: Leveraging Hierarchical Topological Granularity from Cellular Complexes for Graph Contrastive Learning

Related

• Attending to Topological Spaces: The Cellular Transformer
• HTG-GCL: Leveraging Hierarchical Topological Granularity from Cellular Complexes for Graph Contrastive Learning
• From Latent Graph to Latent Topology Inference: Differentiable Cell Complex Module
• Topological Deep Learning: Going Beyond Graph Data
• Topological Neural Networks: Mitigating the Bottlenecks of Graph Neural Networks via Higher-Order Interactions