Adaptive-Step Graph Meta-Learner for Few-Shot Graph Classification

Ma, Ning; Bu, Jiajun; Yang, Jieyu; Zhang, Zhen; Yao, Chengwei; Yu, Zhi; Zhou, Sheng; Yan, Xifeng

doi:10.1145/3340531.3411951

Cited by 23 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meta-Learning on Graph. Meta-learning has been adopted over graph data to deal with various graph learning tasks, including node classification [19,47], link prediction [5,19], graph classification [7,29] and graph alignment [43,46]. A brief survey that summarizes the applications and methods can be found in [30].…”

Section: Related Workmentioning

confidence: 99%

Community Search: A Meta-Learning Approach

Fang¹,

Zhao²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

Community Search (CS) is one of the fundamental graph analysis tasks, which is a building block of various real applications. Given any query nodes, CS aims to find cohesive subgraphs that query nodes belong to. Recently, a large number of CS algorithms are designed. These algorithms adopt pre-defined subgraph patterns to model the communities, which cannot find communities that do not have such pre-defined patterns in real-world graphs. Thereby, machine learning based approaches are proposed to capture flexible community structures by learning from community ground-truth in a data-driven fashion. However, existing approaches rely on sufficient training data to provide enough generalization for machine learning models.In this paper, we study ML-based approaches for community search, under the circumstance that the training data is scarce. To learn from small data, we extract prior knowledge which is shared across different graphs as CS tasks in advance. Subsequent small training data from a new CS task are combined with the learned prior knowledge to help the model well adapt to that specific task. A novel meta-learning based framework, called CGNP, is designed and implemented to fulfill this learning procedure. A meta CGNP model is a task-common node embedding function for clustering by nature, learned by metric-based learning. To the best of our knowledge, CGNP is the first meta model solution for CS. We compare CGNP with traditional CS algorithms, e.g., CTC, ATC, ACQ, and ML baselines on real graph datasets with groundtruth. Our experiments show that CGNP outperforms the native graph algorithms and ML baselines 147% and 113% on F1-score by average.

show abstract

Section: Related Workmentioning

confidence: 99%

Community Search: A Meta-Learning Approach

Fang¹,

Zhao²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In the field of graphs, several recent works propose to conduct graph-based tasks under the few-shot learning scenario [4,19,37]. Among them, GPN [6] proposes to leverage node importance based on Prototypical Networks [28] for better performance, where nodes are classified via finding the nearest class prototype.…”

Section: Few-shot Learning On Graphsmentioning

confidence: 99%

Task-Adaptive Few-shot Node Classification

Wang¹,

Ding²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Node classification is of great importance among various graph mining tasks. In practice, real-world graphs generally follow the long-tail distribution, where a large number of classes only consist of limited labeled nodes. Although Graph Neural Networks (GNNs) have achieved significant improvements in node classification, their performance decreases substantially in such a few-shot scenario. The main reason can be attributed to the vast generalization gap between meta-training and meta-test due to the task variance caused by different node/class distributions in meta-tasks (i.e., node-level and class-level variance). Therefore, to effectively alleviate the impact of task variance, we propose a task-adaptive node classification framework under the few-shot learning setting. Specifically, we first accumulate meta-knowledge across classes with abundant labeled nodes. Then we transfer such knowledge to the classes with limited labeled nodes via our proposed task-adaptive modules. In particular, to accommodate the different node/class distributions among meta-tasks, we propose three essential modules to perform node-level, class-level, and task-level adaptations in each meta-task, respectively. In this way, our framework can conduct adaptations to different meta-tasks and thus advance the model generalization performance on meta-test tasks. Extensive experiments on four prevalent node classification datasets demonstrate the superiority of our framework over the state-of-the-art baselines. Our code is provided at https://github.com/SongW-SW/TENT. CCS CONCEPTS• Computing methodologies → Transfer learning.

show abstract

“…MME proposed minimax entropy loss for the SSDA problem. Based on MME, Meta-MME achieves better generalization with the help of Meta Learning [6,22]. BiAT uses bidirectional adversarial training for generating samples between the source and target domain.…”

Section: Baselinesmentioning

confidence: 99%

Uncertainty-Guided Mixup for Semi-Supervised Domain Adaptation without Source Data

Ma,

Bu,

Zhang

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Present domain adaptation methods usually perform explicit representation alignment by simultaneously accessing the source data and target data. However, the source data are not always available due to the privacy preserving consideration or bandwidth limitation. Source-free domain adaptation aims to solve the above problem by performing domain adaptation without accessing the source data. The adaptation paradigm is receiving more and more attention in recent years, and multiple works have been proposed for unsupervised source-free domain adaptation. However, without utilizing any supervised signal and source data at the adaptation stage, the optimization of the target model is unstable and fragile. To alleviate the problem, we focus on semi-supervised domain adaptation under source-free setting. More specifically, we propose uncertainty-guided Mixup to reduce the representation's intra-domain discrepancy and perform inter-domain alignment without directly accessing the source data. We conduct extensive semi-supervised domain adaptation experiments on various datasets. Our method outperforms the recent semi-supervised baselines and the unsupervised variant also achieves competitive performance. The experiment codes will be released in the future.

show abstract

Adaptive-Step Graph Meta-Learner for Few-Shot Graph Classification

Cited by 23 publications

References 15 publications

Community Search: A Meta-Learning Approach

Community Search: A Meta-Learning Approach

Task-Adaptive Few-shot Node Classification

Uncertainty-Guided Mixup for Semi-Supervised Domain Adaptation without Source Data

Contact Info

Product

Resources

About