Unverified50% confidenceFactExact time
The compound scaling mathematical formulation of EfficientNet is depth d=α^φ, width w=β^φ, and resolution r=γ^φ, where α, β, and γ are determined by grid search on the baseline model B0.
1
Sources
50%
Confidence
Long-term
Relevance
9/5/2026
First Seen
Sources
Related Entities
Related Claims
Unverified大模型底层的数学结构是深度神经网络,本质上是无数个简单函数层层嵌套形成的复合函数65% similarUnverified深度重绘使用MiDaS、DPT等单目深度估计模型从原图中自动推断深度图65% similarUnverified在深度学习中,雅可比矩阵的计算是反向传播算法的数学基础,描述梯度如何在网络层间流动64% similarUnverified模型量化(Quantization)是将深度学习模型参数从高精度浮点数(如FP32、FP16)压缩为低精度表示(如INT8、INT4)的技术,典型方案包括GPTQ、GGUF、AWQ64% similarUnverifiedDeepSeek背后有量化基金幻方量化的自有算力集群支撑64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/859641API
curl https://kongchang.com/api/v1/knowledge/claims/859641MCP
get_claim(id=859641)