Unverified50% confidenceSolutionExact time
The Gaussian distribution PDF contains the term exp(-x²/2σ²), which underflows to zero when x is far from the mean or σ is extremely small; therefore, in practice the log-PDF is commonly computed in log space to ensure numerical stability.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe exponential distribution Exp(λ) describes the waiting time between successive events in a Poisson process and possesses the memoryless property; the Gamma distribution is the distribution of the sum of k independent exponential random variables and degenerates to the exponential distribution when k=1.62% similarUnverifiedScore函数定义为对数密度梯度 ∇x log p(x)，指向概率密度增大最快的方向，其对数梯度使归一化常数消失59% similarUnverified逻辑回归中的交叉熵损失如果直接对 log(0) 求值会产生负无穷，需要加入极小的 epsilon 值或使用 log-sum-exp 技巧58% similarUnverified从信息论角度，DSL约束的本质是降低了输出分布的熵，一个等概率的N选1分布熵值为log₂N57% similarUnverifiedShannon's information entropy formula is H = -Σ p(x) log₂ p(x), which describes the average uncertainty of a random variable.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/831824API
curl https://kongchang.com/api/v1/knowledge/claims/831824MCP
get_claim(id=831824)