Revisiting Concentration of Missing Mass
arXiv:2005.10018v3
AI Analysis
This work addresses a theoretical problem in probability and statistics, but it is incremental as it builds directly on prior results.
The paper tackles the problem of missing mass concentration by developing a new method based on Rosenthal's inequality, resulting in a slight improvement over existing state-of-the-art bounds and simplified proofs.
We revisit the problem of \emph{missing mass concentration}, developing a new method of estimating concentration of heterogenic sums, in spirit of celebrated Rosenthal's inequality. As a result we slightly improve the state-of-art bounds due to Ben-Hamou at al., and simplify the proofs.