Unverified50% confidenceFactExact time
监督学习的目标是学习输入特征X到输出标签Y的映射函数，分为分类（离散标签）和回归（连续标签）两大子任务
1
Sources
50%
Confidence
Long-term
Relevance
7/11/2026
First Seen
Sources
K-means vs 随机森林：如何从根本上判断算法选型？
redditr/learnmachinelearning7/10/2026
Related Claims
Partially VerifiedThe goal of supervised learning is to find a function f that, for a given input x, can predict the corresponding output y.73% similarUnverified回归、分类、聚类构成传统机器学习的基本任务类型：回归预测连续数值，分类将输入归入离散类别，聚类是无监督学习在无标签情况下发现数据自然分组69% similarUnverified成对数据（转换前位图与转换后矢量图一一对应）在监督学习中被称为标注数据，是训练AI模型理想的数据形态68% similarUnverifiedLLM在「生成」和「评估」时激活的推理路径不同，生成任务偏向延续已有语义方向，评估任务激活更多批判性、面向异常检测的推理路径67% similarUnverified神经网络训练的目标是根据已知的一组x和y猜出所有的w和b参数67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/480310API
curl https://kongchang.com/api/v1/knowledge/claims/480310MCP
get_claim(id=480310)