Few-Shot Learning with Graph Neural Networks

2017/11/10 by Víctor García, Joan Bruna, Garcia, Victor +1 · 53 citations
Computer Science · #Domain Adaptation and Few-Shot Learning #Machine Learning and Algorithms #Advanced Graph Neural Networks

paper · pdf · doi:10.48550/arxiv.1711.04043

Abstract

We propose to study the problem of few-shot learning with the prism of inference on a partially observed graphical model, constructed from a collection of input images whose label can be either observed or not. By assimilating generic message-passing inference algorithms with their neural-network counterparts, we define a graph neural network architecture that generalizes several of the recently proposed few-shot learning models. Besides providing improved numerical performance, our framework is easily extended to variants of few-shot learning, such as semi-supervised or active learning, demonstrating the ability of graph-based models to operate well on 'relational' tasks.

Citations

• Gated Graph Sequence Neural Networks

Cited by

• Advancing Cache-Based Few-Shot Classification via Patch-Driven Relational Gated Graph Attention
• Neural Coherence : Find higher performance to out-of-distribution tasks from few samples
• Free Lunch to Meet the Gap: Intermediate Domain Reconstruction for Cross-Domain Few-Shot Learning
• Edge-labeling Graph Neural Network for Few-shot Learning
• Neural Relational Inference for Interacting Systems
• Graph convolutional networks for learning with few clean and many noisy labels
• Graph neural networks: A review of methods and applications
• Leveraging the Feature Distribution in Transfer-based Few-Shot Learning
• Large Margin Few-Shot Learning
• Sequential Graph Convolutional Network for Active Learning
• Adversarial Feature Hallucination Networks for Few-Shot Learning
• A Meta-Learning Approach for Graph Representation Learning in Multi-Task Settings
• Learning Implicit Temporal Alignment for Few-shot Video Classification
• Region Comparison Network for Interpretable Few-shot Image Classification
• Adaptive Prototypical Networks with Label Words and Joint Representation Learning for Few-Shot Relation Classification
• Graph Neural Networks: A Review of Methods and Applications
• Machine Learning on Graphs: A Model and Comprehensive Taxonomy
• Learning Nonparametric Human Mesh Reconstruction from a Single Image without Ground Truth Meshes
• OR-Net: Pointwise Relational Inference for Data Completion under Partial Observation
• Analytical Survey of Learning with Low-Resource Data: From Analysis to Investigation
• Multi-Label Few-Shot Learning for Aspect Category Detection
• Few-shot Relation Extraction via Bayesian Meta-learning on Relation Graphs
• VT-FSL: Bridging Vision and Text with LLMs for Few-Shot Learning
• Adaptive Knowledge-Enhanced Bayesian Meta-Learning for Few-shot Event Detection
• DPGN: Distribution Propagation Graph Network for Few-shot Learning
• Large-Scale Historical Watermark Recognition: dataset and a new consistency-based approach
• DeepEMD: Differentiable Earth Mover's Distance for Few-Shot Learning
• MM-FSOD: Meta and metric integrated few-shot object detection
• Few-shot learning via tensor hallucination
• AGL: a Scalable System for Industrial-purpose Graph Machine Learning
• Weakly-supervised Object Localization for Few-shot Learning and Fine-grained Few-shot Learning
• A Closer Look at Few-shot Classification
• Automated Relational Meta-learning
• Adaptive Deep Kernel Learning
• Image Deformation Meta-Networks for One-Shot Learning
• Improving Few-shot Learning by Spatially-aware Matching and CrossTransformer
• Entity Concept-enhanced Few-shot Relation Extraction
• Modality to Modality Translation: An Adversarial Representation Learning and Graph Fusion Network for Multimodal Fusion
• Meta-learning with differentiable closed-form solvers
• Self-Denoising Neural Networks for Few Shot Learning
• Graphite: Iterative Generative Modeling of Graphs
• Tailoring: encoding inductive biases by optimizing unsupervised\n objectives at prediction time
• Dual Graph Convolutional Networks with Transformer and Curriculum Learning for Image Captioning
• Color as the Impetus: Transforming Few-Shot Learner
• MetaLab: Few-Shot Game Changer for Image Recognition
• Shallow Deep Learning Can Still Excel in Fine-Grained Few-Shot Learning
• FGN: Fully Guided Network for Few-Shot Instance Segmentation
• Induction Networks for Few-Shot Text Classification
• Are Fewer Labels Possible for Few-shot Learning?
• Negative Margin Matters: Understanding Margin in Few-shot Classification
• Learning to Affiliate: Mutual Centralized Learning for Few-shot Classification
• Rethinking Class Relations: Absolute-relative Supervised and Unsupervised Few-shot Learning
• Ensemble Making Few-Shot Learning Stronger

Related

• Graph Few-shot Learning via Knowledge Transfer
• Optimization of Image Embeddings for Few Shot Learning
• Graph-based Interpolation of Feature Vectors for Accurate Few-Shot Classification
• Edge-labeling Graph Neural Network for Few-shot Learning
• Hybrid Graph Neural Networks for Few-Shot Learning