Daily update | 21 June, 2022
StackExchange
Can an irrelevant variable be significant in a regression model?

Source: stats
Views: 351
Score: 6
Tags: regression statistical-significance least-squares econometrics
Why symmetric trees are not used in xgboost

Source: stats
Views: 115
Score: 5
Tags: boosting cart
Why is { w | |w| mod 3 = #_a(w) mod 3 } a Regular Language?

Source: cs
Views: 362
Score: 2
Tags: formal-languages regular-languages automata finite-automata context-free
Using k-means to create labels for supervised learning

Source: datascience
Views: 23
Score: 2
Tags: clustering unsupervised-learning k-means supervised-learning labels
Maximum Likelihood Estimator for Bernoulli distribution

Source: stats
Views: 26
Score: 1
Tags: maximum-likelihood inference likelihood