Search Improves Label for Active Learning
This work addresses active learning efficiency for machine learning practitioners by providing a novel approach that leverages human search capabilities, though it is incremental in enhancing existing methods.
The paper tackles the problem of active learning by introducing a Search oracle alongside the standard Label oracle, showing that this combination yields exponentially large problem-dependent improvements over using Label alone.
We investigate active learning with access to two distinct oracles: Label (which is standard) and Search (which is not). The Search oracle models the situation where a human searches a database to seed or counterexample an existing solution. Search is stronger than Label while being natural to implement in many situations. We show that an algorithm using both oracles can provide exponentially large problem-dependent improvements over Label alone.