NPID (Non-Parametric Instance Discrimination) is a self-supervision approach that takes a non-parametric classification approach. Noise contrastive estimation is used to learn representations. Specifically, distances (similarity) between instances are calculated directly from the features in a non-parametric way.
Source: Unsupervised Feature Learning via Non-Parametric Instance DiscriminationPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Object Detection | 3 | 15.79% |
Semi-Supervised Image Classification | 3 | 15.79% |
Self-Supervised Learning | 2 | 10.53% |
Self-Supervised Image Classification | 2 | 10.53% |
Classifier calibration | 1 | 5.26% |
Few-Shot Learning | 1 | 5.26% |
Fine-Grained Image Recognition | 1 | 5.26% |
Image Classification | 1 | 5.26% |
Semantic Segmentation | 1 | 5.26% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |