Unverified50% confidenceFactExact time
The gradient descent update rule is v → v − η∇C; since η > 0 and ‖∇C‖² ≥ 0, each update is guaranteed to decrease the cost, ΔC ≤ 0.
1
Sources
50%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
梯度下降与反向传播详解：神经网络学习机制从零拆解
Related Claims
UnverifiedGradient descent updates parameters through the iterative formula θ = θ - α∇J(θ), where α is the learning rate and ∇J(θ) is the vector of partial derivatives of the loss function with respect to the parameters.80% similarUnverifiedResearch by Gunasekar et al. in 2017 proved that for linear models, gradient descent is equivalent to minimum-norm interpolation.63% similarUnverified损失函数中 tanh(β·E1·E2) 取 β=10，当 β 足够大时将乘积符号信息锐化为接近 ±1 的值63% similarUnverifiedLinear regression uses mean squared error as the loss function, with the formula L = (Ŷ - Y)².61% similarUnverifiedThe mean squared error (MSE) cost function is defined as C(w,b) = 1/(2n) Σ ‖yᵢ − ŷᵢ‖², where each squared error term is always non-negative.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/556682API
curl https://kongchang.com/api/v1/knowledge/claims/556682MCP
get_claim(id=556682)