Unverified50% confidenceFactExact time
FGSM (Fast Gradient Sign Method) is an adversarial example generation algorithm proposed by Ian Goodfellow in 2014, with the mathematical expression x_adv = x + ε·sign(∇_x L(θ, x, y)).
1
Sources
50%
Confidence
Long-term
Relevance
8/6/2026
First Seen
Sources
ML安全工程师职业规划：从零开始的完整学习路径
Related Claims
Unverified导数的数值近似公式为 (f(a+h)-f(a))/h，源自微积分中极限的定义58% similarUnverified龙曲线的经典 L-系统定义为起始公理 FX，重写规则 X → X+YF+ 和 Y → -FX-Y，其中 F 表示前进画线，+ 表示右转 90°，- 表示左转 90°57% similarPartially VerifiedThe mathematical form of KL divergence is D_KL(p||q) = ∑ p(x_i) log(p(x_i)/q(x_i)).56% similarUnverified球谐函数Y_l^m(θ,φ)是拉普拉斯方程在球坐标下的角度部分解，由阶数l和次数m标定，每个阶数l对应2l+1个函数56% similarUnverifiedThe residual block formula is y = x + F(x); during backpropagation, the gradient passes through the "1" in (1 + dF/dx), providing a shortcut for gradients to flow directly to earlier layers.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/697603API
curl https://kongchang.com/api/v1/knowledge/claims/697603MCP
get_claim(id=697603)