English

An Adaptive Algorithm for Learning with Unknown Distribution Drift

Machine Learning 2023-10-31 v3

Abstract

We develop and analyze a general technique for learning with an unknown distribution drift. Given a sequence of independent observations from the last TT steps of a drifting distribution, our algorithm agnostically learns a family of functions with respect to the current distribution at time TT. Unlike previous work, our technique does not require prior knowledge about the magnitude of the drift. Instead, the algorithm adapts to the sample data. Without explicitly estimating the drift, the algorithm learns a family of functions with almost the same error as a learning algorithm that knows the magnitude of the drift in advance. Furthermore, since our algorithm adapts to the data, it can guarantee a better learning error than an algorithm that relies on loose bounds on the drift. We demonstrate the application of our technique in two fundamental learning scenarios: binary classification and linear regression.

Keywords

Cite

@article{arxiv.2305.02252,
  title  = {An Adaptive Algorithm for Learning with Unknown Distribution Drift},
  author = {Alessio Mazzetto and Eli Upfal},
  journal= {arXiv preprint arXiv:2305.02252},
  year   = {2023}
}

Comments

Updated version for Camera-ready with minor changes in text for readability, and including a new small section on linear regression