Unverified50% confidenceFactExact time
Kaiming He's team pushed the ResNet network depth to 152 layers and resolved the degradation problem through residual connections.
1
Sources
50%
Confidence
Long-term
Relevance
8/22/2026
First Seen
Sources
ResNet残差连接为何有效？深层网络退化问题实验复现
Related Entities
Related Claims
VerifiedResNet (Residual Network) was proposed by Microsoft Research's He Kaiming team in 2015 and solved the vanishing gradient problem in deep network training by introducing residual connections.77% similarUnverifiedResNet（残差网络）通过跳跃连接解决了深层网络的梯度消失问题，能提取层次化视觉特征76% similarUnverifiedResNet（2015，He 等人）通过引入跳跃连接构造恒等映射的学习旁路，将可训练网络深度推进至 152 层乃至千层，并在 ImageNet 上以显著优势夺冠74% similarUnverified残差连接与层归一化解决了深层网络训练不稳定的问题，使模型可堆叠到数十层69% similarUnverifiedResidual Connections add the input of each layer directly to its output, solving the vanishing gradient problem in deep networks66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/788583API
curl https://kongchang.com/api/v1/knowledge/claims/788583MCP
get_claim(id=788583)