Combining active learning suggestions

Alasdair Tran; Cheng Soon Ong; Christian Wolf

doi:10.7717/peerj-cs.157

Combining active learning suggestions

PeerJ Comput Sci. 2018 Jul 23:4:e157. doi: 10.7717/peerj-cs.157. eCollection 2018.

Authors

Alasdair Tran^{1

2}, Cheng Soon Ong^{1

3}, Christian Wolf^{4

5}

Affiliations

¹ Research School of Computer Science, Australian National University, Canberra, ACT, Australia.
² Data to Decisions Cooperative Research Centre, Adelaide, SA, Australia.
³ Machine Learning Research Group, Data61, CSIRO, Canberra, ACT, Australia.
⁴ Research School of Astronomy and Astrophysics, Australian National University, Canberra, ACT, Australia.
⁵ ARC Centre of Excellence for All-sky Astrophysics (CAASTRO), Sydney, NSW, Australia.

Abstract

We study the problem of combining active learning suggestions to identify informative training examples by empirically comparing methods on benchmark datasets. Many active learning heuristics for classification problems have been proposed to help us pick which instance to annotate next. But what is the optimal heuristic for a particular source of data? Motivated by the success of methods that combine predictors, we combine active learners with bandit algorithms and rank aggregation methods. We demonstrate that a combination of active learners outperforms passive learning in large benchmark datasets and removes the need to pick a particular active learner a priori. We discuss challenges to finding good rewards for bandit approaches and show that rank aggregation performs well.

Keywords: Active learning; Bandit; Benchmark; Multiclass classification; Rank aggregation.

Grants and funding

The research was supported by the Data to Decisions Cooperative Research Centre whose activities are funded by the Australian Commonwealth Government’s Cooperative Research Centres Programme. This research was supported by the Australian Research Council Centre of Excellence for All-sky Astrophysics (CAASTRO), through project number CE110001020. The SDSS dataset was extracted from Data Release 12 of SDSS-III. Funding for SDSS-III has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, and the U.S. Department of Energy Office of Science. The SDSS-III web site is http://www.sdss3.org/. SDSS-III is managed by the Astrophysical Research Consortium for the Participating Institutions of the SDSS-III Collaboration including the University of Arizona, the Brazilian Participation Group, Brookhaven National Laboratory, Carnegie Mellon University, University of Florida, the French Participation Group, the German Participation Group, Harvard University, the Instituto de Astrofisica de Canarias, the Michigan State/Notre Dame/JINA Participation Group, Johns Hopkins University, Lawrence Berkeley National Laboratory, Max Planck Institute for Astrophysics, Max Planck Institute for Extraterrestrial Physics, New Mexico State University, New York University, Ohio State University, Pennsylvania State University, University of Portsmouth, Princeton University, the Spanish Participation Group, University of Tokyo, University of Utah, Vanderbilt University, University of Virginia, University of Washington, and Yale University. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.