Abstract
Abstract Even though numerous works focus on the few-shot learning issue by combining meta-learning, there are still limits to traditional graph classification problems. The antecedent algorithms directly extract features from the samples, and do not take into account the preference of the trained model to the previously “seen” targets. In order to overcome the aforementioned issues, an effective strategy with training an unbiased meta-learning algorithm was developed in this paper, which sorted out problems of target preference and few-shot under the meta-learning paradigm. First, the interactive attention extraction module as a supplement to feature extraction was employed, which improved the separability of feature vectors, reduced the preference of the model for a certain target, and remarkably improved the generalization ability of the model on the new task. Second, the graph neural network was used to fully mine the relationship between samples to constitute graph structures and complete image classification tasks at a node level, which greatly enhanced the accuracy of classification. A series of experimental studies were conducted to validate the proposed methodology, where the few-shot and semisupervised learning problem has been effectively solved. It also proved that our model has better accuracy than traditional classification methods on real-world datasets.
Highlights
Even though the algorithms based on deep learning have powerful feature extraction and knowledge expression abilities (Silver et al, 2016; Li et al, 2017; Devlin et al, 2018; Cai et al, 2021), there are still challenges (Li et al, 2019a; Deng et al, 2021a)
In order to alleviate this deviation, this paper proposes an interactive attention extraction module in the feature extraction process, which has been exploited to improve the separability of feature vectors and reduce the model’s preference for a certain target
graph neural networks (GNNs) (Garcia & Bruna, 2017) and EGNN (Kim et al, 2019): In order to prove the advanced nature of our feature extraction module, we compared the GNN based on the non-Euclidean domain proposed by Garcia V et al and the EGNN proposed by Jongmin Kim et al.; they provided an algorithm that uses GNN to solve few-shot learning, and utilize graph model to calculate the relationship between images
Summary
Even though the algorithms based on deep learning have powerful feature extraction and knowledge expression abilities (Silver et al, 2016; Li et al, 2017; Devlin et al, 2018; Cai et al, 2021), there are still challenges (Li et al, 2019a; Deng et al, 2021a). Deep learning methods fully depend on datasets that require considerable labeled samples for training purposes. Deep learning is merely for a specific task. Owing to lacking training samples, the trained models may work well on the training set but encounter parameters overfitting problems on the testing set. Another key point of solving the few-shot learning problem is to overcome the problems of overfitting. Due to these inherent defects of deep learning, it remains gravely difficult to realize artificial intelligence by using deep learning alone
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
