From Cutting Planes Algorithms to Compression Schemes and Active Learning
This work provides a novel framework for converting existing passive learning methods into active learning algorithms, which could benefit researchers in machine learning by offering new tools for efficient data usage.
The paper demonstrates that cutting-plane methods, traditionally used for optimization, can be applied to machine learning tasks such as learning sparse classifiers and designing compression schemes, and can be easily adapted into effective active learning algorithms, with numerical simulations showing their relevance.
Cutting-plane methods are well-studied localization(and optimization) algorithms. We show that they provide a natural framework to perform machinelearning ---and not just to solve optimization problems posed by machinelearning--- in addition to their intended optimization use. In particular, theyallow one to learn sparse classifiers and provide good compression schemes.Moreover, we show that very little effort is required to turn them intoeffective active learning methods. This last property provides a generic way todesign a whole family of active learning algorithms from existing passivemethods. We present numerical simulations testifying of the relevance ofcutting-plane methods for passive and active learning tasks.