An Introduction to ResNets
SMRTR summary
ResNets revolutionized deep learning by enabling training of much deeper neural networks. They solve the degradation problem where deeper networks performed worse, and the shattering gradients problem where gradients become chaotic in very deep networks. ResNets use skip connections to maintain gradient similarity across layers, allowing effective training of extremely deep architectures that outperform shallower networks in computer vision tasks.
SMRTR provides this summary for quick context. The original article belongs to Medium.
Read the original article