ConceptData Heterogeneity / Non-IID

Data Heterogeneity

The phenomenon in federated learning in which data distributions differ across clients, causing the aggregated model to favor certain clients and affecting overall performance and fairness.

Timeline (last 90 days)

Oct 20

Data heterogeneity in federated settings is often called the Non-IID problem, i.e., the data distributions of different clients are inconsistent.

Unverified50%

All Facts (1)

Unverified

Data heterogeneity in federated settings is often called the Non-IID problem, i.e., the data distributions of different clients are inconsistent.

50%

Source Articles

PCHMR Benchmark: Federated Learning for Human Mesh Recovery Under Privacy ConstraintsOct 20