Discovering General-Purpose Active Learning Strategies

Ksenia Konyushkova; Raphael Sznitman; Pascal Fua

Discovering General-Purpose Active Learning Strategies

Machine Learning 2019-04-03 v2 Machine Learning

Authors: Ksenia Konyushkova , Raphael Sznitman , Pascal Fua

Abstract

We propose a general-purpose approach to discovering active learning (AL) strategies from data. These strategies are transferable from one domain to another and can be used in conjunction with many machine learning models. To this end, we formalize the annotation process as a Markov decision process, design universal state and action spaces and introduce a new reward function that precisely model the AL objective of minimizing the annotation cost. We seek to find an optimal (non-myopic) AL strategy using reinforcement learning. We evaluate the learned strategies on multiple unrelated domains and show that they consistently outperform state-of-the-art baselines.

Keywords

active learning reinforcement learning multi-agent reinforcement learning

Cite

@article{arxiv.1810.04114,
  title  = {Discovering General-Purpose Active Learning Strategies},
  author = {Ksenia Konyushkova and Raphael Sznitman and Pascal Fua},
  journal= {arXiv preprint arXiv:1810.04114},
  year   = {2019}
}

Related papers

View all related →