Rademacher complexity is a concept from statistical learning theory that measures the capacity of a class of functions or hypotheses in terms of their ability to fit random noise. Specifically, it quantifies how well a hypothesis class can "respond" to random labels.
Articles by others on the same topic
For independent and identically distributed random variables and independent Rademacher signs , the Rademacher complexity of a function class isIt measures how strongly the class can correlate with independent random signs on a random sample.