Unsupervised Image Classification with Adaptive Nearest Neighbor Selection and Cluster Ensembles

Open paper
Benchmark Model Rank Results
image-clustering-on-cifar-10ICCE Ensemble–Accuracy: 0.9931NMI: 0.9803ARI: 0.9848
image-clustering-on-cifar-10ICCE Ensemble + Self-Training–Accuracy: 0.992NMI: 0.9823ARI: 0.9772
image-clustering-on-cifar-10ICCE stage-1–Accuracy: 0.9927NMI: 0.9793ARI: 0.9838
image-clustering-on-cifar-100ICCE Ensemble–Accuracy: 0.7367NMI: 0.7661
image-clustering-on-cifar-100ICCE Ensemble + Self-Training–Accuracy: 0.7423NMI: 0.7649
image-clustering-on-cifar-100ICCE stage-1–Accuracy: 0.6771NMI: 0.7409
image-clustering-on-stl-10ICCE Ensemble–Accuracy: 0.996NMI: 0.9898ARI: 0.9912
image-clustering-on-stl-10ICCE Ensemble + Self-Training–Accuracy: 0.9958NMI: 0.9906ARI: 0.9888
image-clustering-on-stl-10ICCE stage-1–Accuracy: 0.9957NMI: 0.989ARI: 0.9905
unsupervised-image-classification-on-cifar-10ICCE Ensemble + Self-Training–Accuracy: 99.20
unsupervised-image-classification-on-cifar-10ICCE stage-1–Accuracy: 99.27
unsupervised-image-classification-on-cifar-20ICCE Ensemble + Self-Training–Accuracy: 74.23
unsupervised-image-classification-on-cifar-20ICCE stage-1–Accuracy: 67.71

Rank counts only results with a code link.