Skip Connections

Residual Connection

Introduced by He et al. in Deep Residual Learning for Image Recognition

Residual Connections are a type of skip-connection that learn residual functions with reference to the layer inputs, instead of learning unreferenced functions.

Formally, denoting the desired underlying mapping as $\mathcal{H}({x})$, we let the stacked nonlinear layers fit another mapping of $\mathcal{F}({x}):=\mathcal{H}({x})-{x}$. The original mapping is recast into $\mathcal{F}({x})+{x}$.

The intuition is that it is easier to optimize the residual mapping than to optimize the original, unreferenced mapping. To the extreme, if an identity mapping were optimal, it would be easier to push the residual to zero than to fit an identity mapping by a stack of nonlinear layers.

Source: Deep Residual Learning for Image Recognition


Paper Code Results Date Stars


Task Papers Share
Language Modelling 46 7.21%
Image Classification 29 4.55%
Object Detection 27 4.23%
Semantic Segmentation 26 4.08%
Machine Translation 19 2.98%
Sentiment Analysis 16 2.51%
Action Recognition 12 1.88%
Question Answering 11 1.72%
Text Generation 11 1.72%


Component Type
Batch Normalization
Normalization (optional)
Activation Functions (optional)