no code implementations • 8 Mar 2024 • Alessio Mazzetto
To use more samples, we must resort to samples further in the past, and we incur a drift error due to the bias introduced by the change in distribution.
no code implementations • 2 Jun 2023 • Alessio Mazzetto, Reza Esfandiarpoor, Eli Upfal, Stephen H. Bach
In particular, at each step, our algorithm guarantees an estimation of the current accuracies of the weak supervision sources over a window of past observations that minimizes a trade-off between the error due to the variance of the estimation and the error due to the drift.
no code implementations • 5 Feb 2023 • Alessio Mazzetto, Eli Upfal
We study nonparametric density estimation in non-stationary drift settings.
1 code implementation • 25 May 2022 • Alessio Mazzetto, Cristina Menghini, Andrew Yuan, Eli Upfal, Stephen H. Bach
We develop the first non-trivial lower bound on the worst-case error of the best map from attributes to classes for this setting, even with perfect attribute detectors.