Unverified50% confidenceFactExact time
Distribution shift includes three common types: covariate shift (the input distribution changes but the conditional distribution does not), label shift (the label marginal distribution changes) and concept drift (the conditional distribution changes over time).
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
Unverified分布偏移可细分为协变量偏移、标签偏移和概念漂移，在ASR呼叫中心场景中这三种偏移往往同时存在75% similarUnverifiedAmbiguous category definitions and biased example distributions will contaminate the final results of LLM classification.68% similarUnverifiedPan & Yang（2010）提出的迁移学习分类框架将域差异分解为特征空间差异、边缘分布差异和条件分布差异三个维度63% similarUnverified概念漂移在不同社区有不同术语：统计学家称dataset shift或distribution shift，机器学习社区区分covariate shift和posterior drift，时间序列领域使用non-stationarity63% similarUnverifiedConcept drift refers to the mapping relationship between input data and target labels changing over time, which is different from data drift where only the input distribution changes.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1012851API
curl https://kongchang.com/api/v1/knowledge/claims/1012851MCP
get_claim(id=1012851)