Active learning: which example should we label
If labelling is expensive, let the model choose which example to ask about. The gain is real, and so are its limits and traps.
4 steps
225 XP
A free account is needed
Start the lesson →
Sources
- Settles, B. 2009 · Active Learning Literature Survey · Univ. of Wisconsin-Madison TR 1648
- Lewis, D. D. & Gale, W. A. 1994 · A Sequential Algorithm for Training Text Classifiers · SIGIR 1994
- Gal, Y. et al. 2017 · Deep Bayesian Active Learning with Image Data · ICML 2017
- Lowell, D. et al. 2019 · Practical Obstacles to Deploying Active Learning · EMNLP 2019
ML Academy · an interactive machine learning course that runs in your browser ·
All lessons