Seminar: Sebastian Farquhar - University of Oxford

Date September 16, 2022
Author Hrvoje Stojic

Unbiased Active Learning and Testing

Abstract

Active learning is a powerful tool when labelling data is expensive, but it introduces a bias because the training data no longer follows the population distribution. We can, in fact, remove this bias using corrective weights based on importance sampling. This has two main consequences: first, we show that the bias is actually useful for active learning, especially with overparameterized models like neural networks; second, this technique enables active testing---a new way of doing model evaluation with limited data.

Notes

Share
,,

Related articles

Seminar: Laurence Aitchison - University of Bristol

Seminar: Andreas Krause - ETH Zurich

Seminar: Frank Hutter - University of Freiburg

Seminar: Noémie Jaquier - Karlsruhe Institute of Technology

Optimization Engine
    Learn more
Solutions
Insights
Company
Research
©2024 Secondmind Ltd.