Graph Contrastive Learning with Implicit Augmentations
Existing graph contrastive learning methods rely on augmentation techniques based on random perturbations (e.g., randomly adding or dropping edges and nodes). Nevertheless, altering certain edges or nodes can unexpectedly change the graph characteristics, and choosing the optimal perturbing ratio for each dataset requires onerous manual tuning. In this paper, we introduce Implicit Graph Contrastive Learning (iGCL), which utilizes augmentations in the latent space learned from a Variational Graph Auto-Encoder by reconstructing graph topological structure. Importantly, instead of explicitly sampling augmentations from latent distributions, we further propose an upper bound for the expected contrastive loss to improve the efficiency of our learning algorithm. Thus, graph semantics can be preserved within the augmentations in an intelligent way without arbitrary manual design or prior human knowledge. Experimental results on both graph-level and node-level tasks show that the proposed method achieves state-of-the-art performance compared to other benchmarks, where ablation studies in the end demonstrate the effectiveness of modules in iGCL.
Code (1)
Tasks
Contrastive LearningGraph ClassificationNode ClassificationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Architecture Matters: Uncovering Implicit Mechanisms in Graph Contrastive Learning
With the prosperity of contrastive learning for visual representation learning (VCL), it is also adapted to the graph domain and yields promising performance. However, through a systematic study of various graph contrast…
Contrastive LearningGraph ClassificationGraph LearningInductive Bias+2Are Graph Augmentations Necessary? Simple Graph Contrastive Learning for Recommendation
Contrastive learning (CL) recently has spurred a fruitful line of research in the field of recommendation, since its ability to extract self-supervised signals from the raw data is well-aligned with recommender systems' …
Contrastive LearningRecommendation SystemsTINC: Temporally Informed Non-Contrastive Learning for Disease Progression Modeling in Retinal OCT Volumes
Recent contrastive learning methods achieved state-of-the-art in low label regimes. However, the training requires large batch sizes and heavy augmentations to create multiple views of an image. With non-contrastive meth…
Contrastive LearningRetinal OCT Disease ClassificationSelf-Supervised LearningCGCL: Collaborative Graph Contrastive Learning without Handcrafted Graph Data Augmentations
Unsupervised graph representation learning is a non-trivial topic. The success of contrastive methods in the unsupervised representation learning on structured data inspires similar attempts on the graph. Existing graph …
Contrastive LearningData AugmentationGraph Representation LearningRepresentation Learning+1Localized Contrastive Learning on Graphs
Contrastive learning methods based on InfoNCE loss are popular in node representation learning tasks on graph-structured data. However, its reliance on data augmentation and its quadratic computational complexity might l…
Contrastive LearningData AugmentationRepresentation Learning