Generalized Good-Turing improves missing mass estimation
A Painsky - Journal of the American Statistical Association, 2023 - Taylor & Francis
Consider a finite sample from an unknown distribution over a countable alphabet. The
missing mass refers to the probability of symbols that do not appear in the sample …
missing mass refers to the probability of symbols that do not appear in the sample …
A data-driven missing mass estimation framework
A Painsky - 2022 IEEE International Symposium on Information …, 2022 - ieeexplore.ieee.org
Consider a finite sample from an unknown distribution over a countable alphabet. The
missing mass refers to the probability of symbols that do not appear in the sample. Missing …
missing mass refers to the probability of symbols that do not appear in the sample. Missing …
Confidence intervals for unobserved events
A Painsky - arXiv preprint arXiv:2211.03052, 2022 - arxiv.org
Consider a finite sample from an unknown distribution over a countable alphabet.
Unobserved events are alphabet symbols which do not appear in the sample. Estimating the …
Unobserved events are alphabet symbols which do not appear in the sample. Estimating the …
Consistent estimation of small masses in feature sampling
Consider an (observable) random sample of sizenfrom an infinite population of individuals,
each individual being endowed with a finite set of" features" from a collection of features (Fj) …
each individual being endowed with a finite set of" features" from a collection of features (Fj) …
On missing mass variance
M Skorski - arXiv preprint arXiv:2104.07028, 2021 - arxiv.org
The missing mass refers to the probability of elements not observed in a sample, and since
the work of Good and Turing during WWII, has been studied extensively in many areas …
the work of Good and Turing during WWII, has been studied extensively in many areas …
Mean-squared accuracy of good-turing estimator
M Skorski - 2021 IEEE International Symposium on Information …, 2021 - ieeexplore.ieee.org
The brilliant method due to Good and Turing allows for estimating objects not occurring in a
sample. The problem, known under names “sample coverage” or “missing mass” goes back …
sample. The problem, known under names “sample coverage” or “missing mass” goes back …
[PDF][PDF] Bayesian nonparametric prediction: from species to features
In species sampling models, observations represent the species' labels of distinct animals in
a population. Feature sampling models generalize species sampling models by allowing …
a population. Feature sampling models generalize species sampling models by allowing …
[PDF][PDF] On Mean-Squared Error of Good-Turing Estimator
M Skorski - researchgate.net
What percentage of the population is not represented in a sample? This problem, known
under names” sample coverage” or” missing mass”, goes back to the cryptograpic work of …
under names” sample coverage” or” missing mass”, goes back to the cryptograpic work of …
Consistent estimation of the missing mass for feature models
F Ayed, M Battiston, F Camerlenghi… - arXiv preprint arXiv …, 2019 - arxiv.org
Feature models are popular in machine learning and they have been recently used to solve
many unsupervised learning problems. In these models every observation is endowed with …
many unsupervised learning problems. In these models every observation is endowed with …