Learning Active Learning from Data

Abstract

In this paper, we suggest a novel data-driven approach to active learning (AL). The key idea is to train a regressor that predicts the expected error reduction for a candidate sample in a particular learning state. By formulating the query selection procedure as a regression problem we are not restricted to working with existing AL heuristics; instead, we learn strategies based on experience from previous AL outcomes. We show that a strategy can be learnt either from simple synthetic 2D datasets or from a subset of domain-specific data. Our method yields strategies that work well on real data from a wide range of domains.

Cite

Text

Konyushkova et al. "Learning Active Learning from Data." Neural Information Processing Systems, 2017.

Markdown

[Konyushkova et al. "Learning Active Learning from Data." Neural Information Processing Systems, 2017.](https://mlanthology.org/neurips/2017/konyushkova2017neurips-learning/)

BibTeX

@inproceedings{konyushkova2017neurips-learning,
  title     = {{Learning Active Learning from Data}},
  author    = {Konyushkova, Ksenia and Sznitman, Raphael and Fua, Pascal},
  booktitle = {Neural Information Processing Systems},
  year      = {2017},
  pages     = {4225-4235},
  url       = {https://mlanthology.org/neurips/2017/konyushkova2017neurips-learning/}
}