Mitigating Embedding and Class Assignment Mismatch in Unsupervised Image Classification
Unsupervised image classification is a challenging computer vision task. Deep learning-based algorithms have achieved superb results, where the latest approach adopts unified losses from embedding and class assignment processes. Since these processes inherently have different goals, jointly optimizing them may lead to a suboptimal solution. To address this limitation, we propose a novel two-stage algorithm in which an embedding module for pretraining precedes a refining module that concurrently performs embedding and class assignment. Our model outperforms SOTA when tested with multiple datasets, by substantially high accuracy of 81.0% for the CIFAR-10 dataset (i.e., increased by 19.3 percent points), 35.3% accuracy for CIFAR-100-20 (9.6 pp) and 66.5% accuracy for STL-10 (6.9 pp) in unsupervised tasks.
Code (1)
Tasks
ClassificationGeneral Classificationimage-classificationImage ClassificationImage ClusteringUnsupervised Image ClassificationSimilar Papers 제목 키워드 기반
Unsupervised Graph Clustering with Deep Structural Entropy
Research on Graph Structure Learning (GSL) provides key insights for graph-based clustering, yet current methods like Graph Neural Networks (GNNs), Graph Attention Networks (GATs), and contrastive learning often rely hea…
ClusteringContrastive LearningGraph AttentionGraph Clustering+1Domain Mismatch Doesn’t Always Prevent Cross-lingual Transfer Learning
Cross-lingual transfer learning without labeled target language data or parallel text has been surprisingly effective in zero-shot cross-lingual classification, question answering, unsupervised machine translation, etc. …
Bilingual Lexicon InductionCross-Lingual TransferMachine TranslationQuestion Answering+4Domain Mismatch Doesn't Always Prevent Cross-Lingual Transfer Learning
Cross-lingual transfer learning without labeled target language data or parallel text has been surprisingly effective in zero-shot cross-lingual classification, question answering, unsupervised machine translation, etc. …
Bilingual Lexicon InductionCross-Lingual TransferMachine TranslationQuestion Answering+4Cross-domain Speech Recognition with Unsupervised Character-level Distribution Matching
End-to-end automatic speech recognition (ASR) can achieve promising performance with large-scale training data. However, it is known that domain mismatch between training and testing data often leads to a degradation of …
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Cross-device ASRCross-environment ASR+4Unsupervised Deep Metric Learning via Auxiliary Rotation Loss
Deep metric learning is an important area due to its applicability to many domains such as image retrieval and person re-identification. The main drawback of such models is the necessity for labeled data. In this work, w…
ClusteringImage RetrievalMetric LearningPerson Re-Identification+1