Graph convolutional networks for learning with few clean and many noisy labels
arXiv:1910.00324
Abstract
In this work we consider the problem of learning a classifier from noisy labels when a few clean labeled examples are given. The structure of clean and noisy data is modeled by a graph per class and Graph Convolutional Networks (GCN) are used to predict class relevance of noisy examples. For each class, the GCN is treated as a binary classifier, which learns to discriminate clean from noisy examples using a weighted binary cross-entropy loss function. The GCN-inferred "clean" probability is then exploited as a relevance measure. Each noisy example is weighted by its relevance when learning a classifier for the end task. We evaluate our method on an extended version of a few-shot learning problem, where the few clean examples of novel classes are supplemented with additional noisy data. Experimental results show that our GCN-based cleaning process significantly improves the classification accuracy over not cleaning the noisy data, as well as standard few-shot classification where only few clean examples are used.
References in corpus (11)
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- Geometric deep learning: going beyond Euclidean data
- YFCC100M: The New Data in Multimedia Research
- Deep Convolutional Networks on Graph-Structured Data
- Learning to Reweight Examples for Robust Deep Learning
- Classification with Noisy Labels by Importance Reweighting
- Training Deep Neural Networks on Noisy Labels with Bootstrapping
- Learning to Propagate Labels: Transductive Propagation Network for Few-shot Learning
- Training Convolutional Networks with Noisy Labels
- Few-Shot Learning with Graph Neural Networks