Unverified50% confidenceFactExact time
Non-independent and identically distributed (non-IID) data is the most typical source of statistical heterogeneity in federated learning, causing client drift that manifests as convergence oscillation, accuracy degradation, and systematic bias against certain clients.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedRobust aggregation methods may misjudge legitimate updates from normal clients as anomalous under non-independent and identically distributed (non-IID) data, because differences in data distributions among clients themselves cause gradient directions to diverge.77% similarUnverifiedData heterogeneity in federated settings is often called the Non-IID problem, i.e., the data distributions of different clients are inconsistent.76% similarUnverifiedWhen statistical heterogeneity exists in the data distributions of clients, the global model tends to cater to clients with larger data volumes and mainstream distributions, causing clients that deviate from the mainstream to exhibit significantly worse local accuracy.65% similarUnverifiedDomain shift refers to the performance degradation that occurs when the distribution of the model's training data is inconsistent with that of the deployment data.65% similarVerified数据漂移（Data Drift）指生产环境中输入数据的统计分布随时间发生变化，导致模型接收数据与训练时分布不匹配64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1113854API
curl https://kongchang.com/api/v1/knowledge/claims/1113854MCP
get_claim(id=1113854)