Planned intervention: On Wednesday April 3rd 05:30 UTC Zenodo will be unavailable for up to 2-10 minutes to perform a storage cluster upgrade.
Published June 22, 2021 | Version v1
Conference paper Open

Neighborhood Contrastive Learning for Novel Class Discovery

  • 1. University of Trento, Italy
  • 2. University of Trento/FBK, Italy
  • 3. Xiamen University

Description

In this paper, we address Novel Class Discovery (NCD), the task of unveiling new classes in a set of unlabeled samples given a labeled dataset with known classes. We exploit the peculiarities of NCD to build a new framework, named Neighborhood Contrastive Learning (NCL), to learn discriminative representations that are important to clustering performance. Our contribution is twofold. First, we find that a feature extractor trained on the labeled set generates representations in which a generic query sample and
its neighbors are likely to share the same class. We exploit this observation to retrieve and aggregate pseudo-positive pairs with contrastive learning, thus encouraging the model to learn more discriminative representations. Second, we notice that most of the instances are easily discriminated by the network, contributing less to the contrastive loss. To overcome this issue, we propose to generate hard negatives by mixing labeled and unlabeled samples in the feature space. We experimentally demonstrate that these two ingredients significantly contribute to clustering performance and lead our model to outperform state-of-the-art methods by a large margin (e.g., clustering accuracy +13% on CIFAR-100 and +8% on ImageNet).

Notes

IEEE/CVF International Conference on Computer Vision and Pattern Recognition (CVPR'21)

Files

Zhong_Neighborhood_Contrastive_Learning_for_Novel_Class_Discovery_CVPR_2021_paper.pdf

Additional details

Funding

AI4Media – A European Excellence Centre for Media, Society and Democracy 951911
European Commission