Unverified50% confidenceFactExact time
Convolutional Neural Networks (CNNs) have built-in inductive biases of local connectivity and weight sharing, assuming that spatial features possess locality and translation invariance.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe core design principles of CNNs are local receptive fields and weight sharing, which make their parameter count far smaller than fully connected networks and give them inherent translation invariance.79% similarUnverified残差连接（Residual Connection）由何恺明等人在2015年的ResNet论文中提出，用于解决深层网络中的信息损失问题65% similarUnverified边缘-云协同架构强依赖网络连接，离线场景下功能大幅退化65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/895919API
curl https://kongchang.com/api/v1/knowledge/claims/895919MCP
get_claim(id=895919)