Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output by a weight. This means there are $n_{\text{inputs}}*n_{\text{outputs}}$ parameters, which can lead to a lot of parameters for a sizeable network.
$$h_{l} = g\left(\textbf{W}^{T}h_{l-1}\right)$$
where $g$ is an activation function.
Image Source: Deep Learning by Goodfellow, Bengio and Courville
Paper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Language Modelling | 45 | 5.80% |
Retrieval | 32 | 4.12% |
Large Language Model | 27 | 3.48% |
Question Answering | 26 | 3.35% |
Semantic Segmentation | 25 | 3.22% |
Decoder | 23 | 2.96% |
Image Classification | 14 | 1.80% |
Text Generation | 13 | 1.68% |
In-Context Learning | 12 | 1.55% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |