Wiki-CS: A Wikipedia-Based Benchmark for Graph Neural Networks

Péter, Mernyei,; Cangea, Cătălina

doi:10.48550/arxiv.2007.02901

Cited by 47 publications

(52 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further information on the datasets can be found in Table 1. To keep fair, for transductive tasks, we split Amazon-Photo, Amazon-Computers, Wiki-CS and Coauthor-CS for the training, validation and testing following (Zhu et al 2021b;Mernyei and Cangea 2020). For inductive task, we split Reddit and Flickr following (Velickovic et al 2019;Zeng et al 2019).…”

Section: Datasetsmentioning

confidence: 99%

ProGCL: Rethinking Hard Negative Mining in Graph Contrastive Learning

Xia¹,

Wu²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Contrastive learning (CL) has emerged as a dominant technique for unsupervised representation learning which embeds augmented versions of the anchor close to each other (positive samples) and pushes the embeddings of other samples (negative samples) apart. As revealed in recent works, CL can benefit from hard negative samples (negative samples that are difficult to distinguish from the anchor). However, we observe minor improvement or even performance drop when we adopt existing hard negative mining techniques in Graph Contrastive Learning (GCL). We find that many hard negative samples similar to anchor point are false negative ones (samples from the same class as anchor point) in GCL, which is different from CL in computer vision and will lead to unsatisfactory performance of existing hard negative mining techniques in GCL. To eliminate this bias, we propose Debiased Graph Contrastive Learning (DGCL), a novel and effective method to estimate the probability whether each negative sample is true or not. With this probability, we devise two schemes (i.e., DGCL-weight and DGCL-mix) to boost the performance of GCL. Empirically, DGCL outperforms or matches previous unsupervised state-of-the-art results on several benchmarks and even exceeds the performance of supervised ones. The code for reproducing the results can be found in the supplementary materials.

show abstract

Section: Datasetsmentioning

confidence: 99%

ProGCL: Rethinking Hard Negative Mining in Graph Contrastive Learning

Xia¹,

Wu²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…We perform experiments on 7 real-world datasets widely used in the literature [48,22,49], i.e., Physics, CS, Photo, Computers, WikiCS, Citeseer and CoraFull. To demonstrate the effectiveness of the proposed framework, we follow [15] and evaluate all methods on two different downstream tasks: node clustering and node classification.…”

Section: Experimental Settingmentioning

confidence: 99%

Automated Self-Supervised Learning for Graphs

Jin¹,

Liu²,

Zhao³

et al. 2021

Preprint

View full text Add to dashboard Cite

Graph self-supervised learning has gained increasing attention due to its capacity to learn expressive node representations. Many pretext tasks, or loss functions have been designed from distinct perspectives. However, we observe that different pretext tasks affect downstream tasks differently across datasets, which suggests that searching over pretext tasks is crucial for graph self-supervised learning. Different from existing works focusing on designing single pretext tasks, this work aims to investigate how to automatically leverage multiple pretext tasks effectively. Nevertheless, evaluating representations derived from multiple pretext tasks without direct access to ground truth labels makes this problem challenging. To address this obstacle, we make use of a key principle of many real-world graphs, i.e., homophily, or the principle that "like attracts like," as the guidance to effectively search various self-supervised pretext tasks. We provide theoretical understanding and empirical evidence to justify the flexibility of homophily in this search task. Then we propose the AUTOSSL framework which can automatically search over combinations of various self-supervised tasks. By evaluating the framework on 7 real-world datasets, our experimental results show that AUTOSSL can significantly boost the performance on downstream tasks including node clustering and node classification compared with training under individual tasks. Code will be released at https://github.com/ChandlerBang/AutoSSL.

show abstract

“…In this section, we apply our methods on six graph datasets (Table 4) for single-hop node classification using permutationinvariant representation, and compare against state-of-the-art (SOTA) graph neural networks and collective inference methods. The Friendster dataset is a social network graph (Teixeira, Jalaian, and Ribeiro 2019); the Pubmed, Cora and Citeseer are three classic citation network benchmarks (Namata et al 2012); the Arxiv is from the Open Graph Benchmark (OGB) (Hu et al 2020); and wikiCS is a web graph (Mernyei and Cangea 2020).…”

Section: Node Classification Tasksmentioning

confidence: 99%

Set Twister for Single-hop Node Classification

Zhou¹,

Rao²,

Ribeiro³

2021

Preprint

View full text Add to dashboard Cite

Node classification is a central task in relational learning, with the current state-of-the-art hinging on two key principles: (i) predictions are permutation-invariant to the ordering of a node's neighbors, and (ii) predictions are a function of the node's r-hop neighborhood topology and attributes, r ≥ 2. Both graph neural networks and collective inference methods (e.g., belief propagation) rely on information from up to rhops away. In this work, we study if the use of more powerful permutation-invariant functions can sometimes avoid the need for classifiers to collect information beyond 1-hop. Towards this, we introduce a new architecture, the Set Twister, which generalizes DeepSets (Zaheer et al.), a simple and widely-used permutation-invariant representation. Set Twister theoretically increases expressiveness of DeepSets, allowing it to capture higher-order dependencies, while keeping its simplicity and low computational cost. Empirically, we see accuracy improvements of Set Twister over DeepSets as well as a variety of graph neural networks and collective inference schemes in several tasks, while showcasing its implementation simplicity and computational efficiency.

show abstract

Wiki-CS: A Wikipedia-Based Benchmark for Graph Neural Networks

Cited by 47 publications

References 0 publications

ProGCL: Rethinking Hard Negative Mining in Graph Contrastive Learning

ProGCL: Rethinking Hard Negative Mining in Graph Contrastive Learning

Automated Self-Supervised Learning for Graphs

Set Twister for Single-hop Node Classification

Contact Info

Product

Resources

About