Unverified50% confidenceTradeoffExact time
In high-dimensional medical data, assigning higher class weights to minority classes or using adaptive loss functions such as Focal Loss is generally more robust than SMOTE oversampling.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedInfoNCE 损失通过在一批候选中最大化正样本对相似度、最小化负样本对相似度进行训练，温度超参数τ越小分布越集中74% similarUnverified贝叶斯最优分类器在0-1损失下等价于选择后验概率最大的类别，回归问题中条件期望在平方损失下是最优预测68% similarUnverified绝大多数常见损失函数都可以从最大似然估计推导出来68% similarUnverifiedIn the no-show prediction task, the cost of missing truly absent patients (false negatives) is far higher than the cost of misclassifying patients who do show up, so recall should be prioritized.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/896955API
curl https://kongchang.com/api/v1/knowledge/claims/896955MCP
get_claim(id=896955)