ConceptAI & ML
Sampling error
Discrepancies between generated statistical outputs and expected values due to sample size limitations.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Bellman operator1Empirical expectation1Finite sample bounds1Hoeffding's inequality1Probability distribution1Probability samples1Statistical correlation1Total variation distance1
Lecturers
3Each in their own terms, from the easiest level.
NPTEL-NOC IITM
Sampling error represents the discrepancy between sample results and population parameters due to the fact that no sample is a perfect mirror image of the population.Sergey Levine
The error introduced because empirical estimates of the reward and transition functions derived from a finite number of samples differ from the true expected values.
Connections
6The lectures that define it or assume it, and the concepts next to it.