Online Learning with Improving Agents: Multiclass, Budgeted Agents and Bandit Learners
Machine Learning
2026-02-20 v1 Machine Learning
Abstract
We investigate the recently introduced model of learning with improvements, where agents are allowed to make small changes to their feature values to be warranted a more desirable label. We extensively extend previously published results by providing combinatorial dimensions that characterize online learnability in this model, by analyzing the multiclass setup, learnability in a bandit feedback setup, modeling agents' cost for making improvements and more.
Cite
@article{arxiv.2602.17103,
title = {Online Learning with Improving Agents: Multiclass, Budgeted Agents and Bandit Learners},
author = {Sajad Ashkezari and Shai Ben-David},
journal= {arXiv preprint arXiv:2602.17103},
year = {2026}
}