Auxiliary Classifiers are type of architectural component that seek to improve the convergence of very deep networks. They are classifier heads we attach to layers before the end of the network. The motivation is to push useful gradients to the lower layers to make them immediately useful and improve the convergence during training by combatting the vanishing gradient problem. They are notably used in the Inception family of convolutional neural networks.
Paper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Classification | 53 | 8.86% |
General Classification | 47 | 7.86% |
Semantic Segmentation | 43 | 7.19% |
Classification | 41 | 6.86% |
Object Detection | 24 | 4.01% |
Quantization | 18 | 3.01% |
Image Segmentation | 13 | 2.17% |
Image Generation | 12 | 2.01% |
Object Recognition | 11 | 1.84% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |