Vincent Gripon's Homepage

Research and Teaching Blog

Improving Classification Accuracy with Graph Filtering

M. Hamidouche, C. Lassance, Y. Hu, L. Drumetz, B. Pasdeloup and V. Gripon, "Improving Classification Accuracy with Graph Filtering," in ArXiv Preprint, 2021.

In machine learning, classifiers are typically susceptible to noise in the training data. In this work, we aim at reducing intra-class noise with the help of graph filtering to improve the classification performance. Considered graphs are obtained by connecting samples of the training set that belong to a same class depending on the similarity of their representation in a latent space. We show that the proposed graph filtering methodology has the effect of asymptotically reducing intra-class variance, while maintaining the mean. While our approach applies to all classification problems in general, it is particularly useful in few-shot settings, where intra-class noise can have a huge impact due to the small sample selection. Using standardized benchmarks in the field of vision, we empirically demonstrate the ability of the proposed method to slightly improve state-of-the-art results in both cases of few-shot and standard classification.

Download manuscript.

Bibtex
@inproceedings{HamLasHuDruPasGri2021,
  author = {Mounia Hamidouche and Carlos Lassance and
Yuqing Hu and Lucas Drumetz and Bastien Pasdeloup and
Vincent Gripon},
  title = {Improving Classification Accuracy with
Graph Filtering},
  booktitle = {ArXiv Preprint},
  year = {2021},
}




You are the 1338719th visitor

Vincent Gripon's Homepage