Interpretable Image Classification with Differentiable Prototypes Assignment
We introduce ProtoPool, an interpretable image classification model with a pool of prototypes shared by the classes. The training is more straightforward than in the existing methods because it does not require the pruning stage. It is obtained by introducing a fully differentiable assignment of prototypes to particular classes. Moreover, we introduce a novel focal similarity function to focus the model on the rare foreground features. We show that ProtoPool obtains state-of-the-art accuracy on the CUB-200-2011 and the Stanford Cars datasets, substantially reducing the number of prototypes. We provide a theoretical analysis of the method and a user study to show that our prototypes are more distinctive than those obtained with competitive methods.
Code (1)
Tasks
Classificationimage-classificationImage ClassificationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Representation Learning via Consistent Assignment of Views over Random Partitions
We present Consistent Assignment of Views over Random Partitions (CARP), a self-supervised clustering method for representation learning of visual features. CARP learns prototypes in an end-to-end online fashion using gr…
Copy DetectionImage RetrievalLinear evaluationRepresentation Learning+2Consistent Assignment for Representation Learning
We introduce Consistent Assignment for Representation Learning (CARL). An unsupervised learning method to learn visual representations by combining contrastive learning with deep clustering. By viewing contrastive learn…
ClusteringContrastive LearningDeep ClusteringRepresentation LearningProtoSSL: Interpretable Prototype Learning from Unlabeled Time-Series Data
In time-series domains where both predictive performance and interpretability are essential, deep neural networks achieve strong results but provide limited insight into how their predictions are made. Projection-based p…
Audio ClassificationRepresentation Learning via Consistent Assignment of Views to Clusters
We introduce Consistent Assignment for Representation Learning (CARL), an unsupervised learning method to learn visual representations by combining ideas from self-supervised contrastive learning and deep clustering. By …
ClusteringContrastive LearningDeep ClusteringLinear evaluation+2CLAP: Unsupervised 3D Representation Learning for Fusion 3D Perception via Curvature Sampling and Prototype Learning
Unsupervised 3D representation learning via masked-and-reconstruction with differentiable rendering is promising to reduce the labeling burden for fusion 3D perception. However, previous literature conduct pre-training f…
GPURepresentation LearningUnsupervised Pre-training