Learning Active Learning from Data

Konyushkova, Ksenia; Sznitman, Raphael; Fua, Pascal (2017). Learning Active Learning from Data (In Press). In: Conference on Neural Information Processing Systems (NIPS).

Text 1703.03365.pdf - Published Version Restricted to registered users only Available under License Publisher holds Copyright. Download (632kB) | Request a copy

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.

Interest & Impact

Downloads

4 since deposited on 08 Nov 2017
0 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item