Unverified50% confidenceFactExact time
余弦退火让学习率按余弦曲线平滑下降至接近零并周期性重启，被SGDR算法正式化
1
Sources
50%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
Related Claims
Unverified残差结构让F=0成为默认选项，只要F(x)权重接近零残差块即退化为恒等映射，从而解决SGD难以学习恒等映射的优化困境67% similarUnverifiedAdaGrad通过累积历史梯度平方实现自适应，但会导致学习率单调递减直至停止更新64% similarVerified梯度下降是机器学习中最核心的优化算法，学习率过大会导致震荡、过小则收敛缓慢61% similarUnverifiedNeRF训练耗时长、推理速度慢，输出的隐式表示无法直接转化为可编辑的几何网格60% similarUnverified梯度下降的权重更新规则为 w = w - lr * gradient，其中lr是学习率59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/558233API
curl https://kongchang.com/api/v1/knowledge/claims/558233MCP
get_claim(id=558233)