Improving Unsupervised Image Clustering With Robust Learning

Unsupervised image clustering methods often introduce alternative objectives to indirectly train the model and are subject to faulty predictions and overconfident results. To overcome these challenges, the current research proposes an innovative model RUC that is inspired by robust learning. RUC's novelty is at utilizing pseudo-labels of existing image clustering models as a noisy dataset that may include misclassified samples. Its retraining process can revise misaligned knowledge and alleviate the overconfidence problem in predictions. The model's flexible structure makes it possible to be used as an add-on module to other clustering methods and helps them achieve better performance on multiple datasets. Extensive experiments show that the proposed model can adjust the model confidence with better calibration and gain additional robustness against adversarial noise.

PDF Abstract CVPR 2021 PDF CVPR 2021 Abstract

Results from the Paper

 Ranked #1 on Image Clustering on CIFAR-100 (Train Set metric, using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Image Clustering CIFAR-10 RUC Accuracy 0.903 # 4
Backbone ResNet-18 # 1
Unsupervised Image Classification CIFAR-10 RUC Accuracy 90.3 # 2
Image Clustering CIFAR-100 RUC Train Set Train # 1
Unsupervised Image Classification CIFAR-20 RUC Accuracy 54.3 # 5
Image Clustering STL-10 RUC Accuracy 0.867 # 5
Backbone ResNet-18 # 1
Unsupervised Image Classification STL-10 RUC Accuracy 86.7 # 3


No methods listed for this paper. Add relevant methods here