An Unsupervised Learning Classifier with Competitive Error Performance
This addresses the problem of unsupervised classification for computer vision, though it is incremental as it slightly underperforms supervised methods.
The paper tackles unsupervised classification by proposing a model that incrementally adjusts discriminative hyperplanes, achieving 6.2% Top 3 error on an ImageNet subset—only about 2% worse than supervised k-NN with the same features.
An unsupervised learning classification model is described. It achieves classification error probability competitive with that of popular supervised learning classifiers such as SVM or kNN. The model is based on the incremental execution of small step shift and rotation operations upon selected discriminative hyperplanes at the arrival of input samples. When applied, in conjunction with a selected feature extractor, to a subset of the ImageNet dataset benchmark, it yields 6.2 % Top 3 probability of error; this exceeds by merely about 2 % the result achieved by (supervised) k-Nearest Neighbor, both using same feature extractor. This result may also be contrasted with popular unsupervised learning schemes such as k-Means which is shown to be practically useless on same dataset.