Unverified50% confidenceFactExact time
The binary cross-entropy loss formula is -[y·log(p) + (1-y)·log(1-p)], where y is the true label and p is the model's predicted probability for the positive class.
1
Sources
50%
Confidence
Long-term
Relevance
8/10/2026
First Seen
Sources
Related Claims
Unverified交叉熵损失函数鼓励模型将正确类别的logit推向无穷大，导致深度网络预测过于确定69% similarUnverifiedThe Shannon entropy of a random variable is defined as H(X) = -∑ p(x) log p(x), measuring the average uncertainty of the random variable.68% similarUnverifiedPerplexity is defined as PPL = 2^{H(p,q)}, where H(p,q) is the cross-entropy of the model distribution relative to the true distribution, and no model's perplexity can fall below the value corresponding to the true entropy rate of the language.66% similarUnverifiedThe parameters of logistic regression are estimated by maximizing the log-likelihood function (equivalent to minimizing the cross-entropy loss).66% similarUnverified交叉熵恢复率通过将SAE重建激活替代原始激活输入到模型后续层比较交叉熵损失变化来计算，衡量SAE是否保留对下游计算重要的信息65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/721283API
curl https://kongchang.com/api/v1/knowledge/claims/721283MCP
get_claim(id=721283)